---
title: "AI Templates: Deploy Open Source AI Apps in One Click"
description: "Deploy 500+ open source AI templates on Railway in one click. Run Ollama, Open WebUI, vector databases, and agent stacks on always-on infrastructure."
url: https://railway.com/deploy/category/ai
---

# Deploy AI apps, agents, and LLM stacks in one click

Templates in this category are prebuilt AI stacks you can deploy on Railway in one click. Each one spins up its services, databases, volumes, and environment variables in your own project, so an app like Open WebUI or a vector database like pgvector is live in a couple of minutes.

Everything runs on always-on servers with no cold starts, which matters when your app is serving a model, syncing embeddings, or waiting on webhooks. Pricing is usage based and you can start on a free trial.

## Common questions about AI templates

### How do I deploy an AI app on Railway?

Pick a template from this page and click Deploy. Railway creates the services, databases, and environment variables the app needs, then builds and deploys everything in your own project. Most templates are live in a few minutes.

### Can I run open source LLMs like Ollama on Railway?

Yes. The Ollama template deploys in one click, and you can pair it with a chat UI such as Open WebUI or LobeChat. Your models run 24/7 on always-on servers.

### Do AI templates include a vector database?

Many do. Templates such as pgvector and Chroma give you a vector store in one click, and you can add one to any existing project. Reference variables wire the connection details into your app automatically.

### Can I self-host a ChatGPT style chat UI?

Yes. Open WebUI, LobeChat, and AnythingLLM all have one-click templates in this category. You control the deployment, the data, and which model providers it talks to.

### How do I host an AI agent 24/7?

Deploy an agent template and it runs continuously, with no cold starts and no idle shutdowns unless you turn on App Sleeping. With usage based pricing you pay for the resources the agent actually uses.

## Related

- Browse all templates: https://railway.com/deploy
- Deploy with AI coding agents: https://railway.com/agents
- Sandboxes for agents: https://railway.com/sandboxes
- Railway MCP server: https://railway.com/mcp
- Railway docs: https://docs.railway.com

## All 1105 AI templates

- [9Router](https://railway.com/deploy/9router) - Self-hosted AI router for Claude, Gemini, Cursor & OpenAI tools.
- [Hermes Agent by Nous Research](https://railway.com/deploy/hermes-railway-template) - Template for Hermes Agent by Nous Research
- [Hermes](https://railway.com/deploy/hermes-agent-nous-research) - Self Host Hermes Agent. Self Improving Agent By Nous Research.
- [Hermes Agent](https://railway.com/deploy/hermes-agent-ai) - Self Host Hermes Agent. Autonomous AI agent w/ self-improving, tools & more
- [Buzz Relay - Block](https://railway.com/deploy/buzz-relay-block) - Deploy and Host Buzz Relay on Railway
- [Hermes Agent](https://railway.com/deploy/hermes-agent) - The agent that grows with you
- [LibreChat](https://railway.com/deploy/librechat-official) - Open-source ChatGPT clone: multi-model, agents, search, multi-user auth
- [OpenClaw (Clawdbot) - Railway Template](https://railway.com/deploy/clawdbot-railway-template) - Quickly deploy your openclaw (clawdbot) on Railway: https://openclaw.ai
- [OpenClaw](https://railway.com/deploy/openclaw) - [Homebrew support] The AI that actually does things
- [9Router — Self-Hosted AI Gateway for 60+ Providers](https://railway.com/deploy/9router-ai-gateway) - Route Claude Code & Cursor to 60+ LLM providers with fallback
- [Hermes Agent - Official Dashboard](https://railway.com/deploy/hermes-agent-official) - Automate AI agent tasks with an official dashboard and self-hosted runtime.
- [Möbius](https://railway.com/deploy/mobius) - Your portal to the world of AI agents.
- [Buzz - Block](https://railway.com/deploy/buzz) - Code review, chat and git for humans and AI coding agents. By Block.
- [Bytebot](https://railway.com/deploy/bytebot) - AI desktop agent that automates any computer task through natural language
- [9router](https://railway.com/deploy/9router-gateway) - Self-host 9router: OpenAI-compatible endpoint, 100+ models, fallback & more
- [Ollama](https://railway.com/deploy/ollama) - A powerful and user-friendly platform for running LLMs
- [SillyTavern](https://railway.com/deploy/YuyzZp) - LLM Frontend for Power Users
- [Paperclip](https://railway.com/deploy/paperclip) - Deploy Paperclip in one click with persistent volume storage.
- [Open WebUI [with Postgres & Redis]](https://railway.com/deploy/openwebui) - Self-host OpenWebUI; Postgres survives redeploys, Redis scales sessions
- [OmniRoute](https://railway.com/deploy/omniroute-1) - OmniRoute — LLM gateway with provider routing, fallbacks, compression
- [Hermes Agent 🤖](https://railway.com/deploy/hermes-agent-nousresearch) - An OpenClaw alternative with dashboard. Self-host, deploy in 1-click 🚀
- [AnythingLLM](https://railway.com/deploy/anythingllm) - An all-in-one application for RAG, AI Agents, multi-user management & more
- [9Router — AI Gateway](https://railway.com/deploy/9router-railway-template) - Self-hosted AI router for Claude Code, Cursor, Copilot & 60+ AI providers
- [BankMCP](https://railway.com/deploy/bankmcp) - Deploy and Host BankMCP™ on Railway
- [DeepSeek Harness | (Just Updated) DeepSeek's Own Coding Agent, Gated](https://railway.com/deploy/deepseek-harness-or-just-updated-deepsee) - DeepSeek's own coding agent, always on and behind a password.
- [AgentOS](https://railway.com/deploy/agentos-1) - Deploy and run your AI workforce with AgentOS, powered by OpenClaw.
- [OpenCode CLI + Web App](https://railway.com/deploy/opencode-cli-web-app) - OpenCode web UI on a public URL + SSH TUI, all in one Railway box
- [nofx](https://railway.com/deploy/nofx) - AI-Powered Trading Platform
- [OpenCode](https://railway.com/deploy/opencode) - always-on autonomous AI coding agent.
- [Ollama + OpenWebUI](https://railway.com/deploy/ollama-openwebui-or-self-host-your-own-c) - [Jun '26] Deploy open-source LLMs with a simple, private web interface.
- [LiteLLM](https://railway.com/deploy/Lm9gxI) - LiteLLM: LLM router, virtual keys, budgets, cost tracking, guardrails
- [Open WebUI on Railway](https://railway.com/deploy/open-webui-2) - [Jun '26] OpenWebUI | Self-hosted AI interface. Connect with any LLM model.
- [Baalda Server](https://railway.com/deploy/baalda-server) - The local-first Markdown second brain.
- [FreeLLMAPI](https://railway.com/deploy/freellmapi) - One OpenAI endpoint. Sixteen free LLM providers. ~1.7B tokens per month.
- [SearXNG (w/ Official Image)](https://railway.com/deploy/searxng-w-official-i) - Bring powerful AI search to your apps with a self-hosted metasearch engine
- [ElevenLabs Audio MCP for ChatGPT](https://railway.com/deploy/elevenlabs-audio-mcp-for-chatgpt) - Self-hosted ElevenLabs speech MCP. Bring your own API key and voice ID.
- [FlowiseAI [Updated Sep ’26] (Visual LLM Workflow Builder)](https://railway.com/deploy/flowise) - Flowise [Sep ’26], Visual LLM Flow Builder (OpenPipe Alternative), Selfhost
- [New API](https://railway.com/deploy/new-api) - OpenAI-compatible API gateway for multiple LLM providers
- [Open Wearables](https://railway.com/deploy/open-wearables-1) - Self-hosted platform to unify wearable data.
- [9Router](https://railway.com/deploy/9router-2) - Self-hosted AI router for Claude, Gemini, Cursor & OpenAI tools.
- [OpenCode](https://railway.com/deploy/opencode-ai) - AI coding agent that writes and runs code in your project
- [OpenClaw (Clawdbot) - Latest](https://railway.com/deploy/openclaw-clawdbot-latest) - Quickly deploy your openclaw (clawdbot) on Railway: https://openclaw.ai
- [Buzz Relay](https://railway.com/deploy/buzz-relay) - A workspace where humans and agents build together, on a relay you own.
- [Hermes Agent](https://railway.com/deploy/hermes-agent-4) - Self-hosted Hermes Agent with admin UI, dashboard, metrics & backups.
- [Hermes Agent (with Official Dashboard)](https://railway.com/deploy/hermes-agent-with-official-dashboard) - Deploy your own Hermes agent with the official Hermes dashboard.
- [sub2api-railway-template](https://railway.com/deploy/sub2api-railway-template) - Subscription to API Conversion Platform
- [n8n-mcp](https://railway.com/deploy/n8n-mcp) - Deploy n8n-mcp to enable AI agents to build and edit n8n workflows
- [Hermes Agent | OpenClaw Alternative with Dashboard](https://railway.com/deploy/hermes-agent-or-openclaw-alternative-wit) - Self-Hosted Hermes AI Agent for Telegram, Discord & Slack
- [Hermes Agent](https://railway.com/deploy/ai-hermes-agent) - Self-host Hermes Agent, the open-source AI agent from Nous Research
- [Opencode Server](https://railway.com/deploy/opencode-server) - Opencode fully featured server for AI Agents. Web, Server, and CLI mode.
- [CLIProxyAPI - No API Key Needed](https://railway.com/deploy/cliproxyapi-no-api-key-needed) - OpenAI/Gemini/Claude/Codex compatible API proxy for CLI with OAuth support
- [SillyTavern in the Cloud — No Node.js Install. Any Device, Always-On.](https://railway.com/deploy/sillytavern-cloud) - Run SillyTavern in the cloud — no Node.js, no install, any device.
- [Hermes Agent](https://railway.com/deploy/hermes-agent-railway) - Run Hermes Agent on Railway
- [AlphaClaw Quick Start](https://railway.com/deploy/openclaw-fast-start) - The ultimate OpenClaw harness. Deploy in minutes, up for months.
- [Langflow](https://railway.com/deploy/JMXEWp) - Simple Langflow deployment using PostgreSQL as the database.
- [Dify](https://railway.com/deploy/dify) - An open-source LLM app development platform
- [9Router — Self-Hosted OpenRouter Alternative](https://railway.com/deploy/9router-openrouter-alternative) - Zero-markup OpenRouter alternative — self-host 60+ LLM providers
- [Buzz by Block](https://railway.com/deploy/buzz-by-block) - Deploy and Host Buzz Relay on Railway
- [SillyTavern](https://railway.com/deploy/sillytavern-llm-frontend) - Power-user AI character chat — any LLM, uncensored, your server
- [LibreChat | Self-Hosted AI Chat with Multi-Provider Support](https://railway.com/deploy/librechat) - 1-Click LibreChat deploy (OpenAI, Anthropic, Google, Ollama, RAG API)
- [Coder](https://railway.com/deploy/coder) - Secure environments for developers and their agents.
- [LobeChat | Multi-Model AI/LLM Support](https://railway.com/deploy/lobechat) - Self Host Lobechat. Private AI chat with 40+ providers, RAG, plugins & more
- [Buzz](https://railway.com/deploy/buzz-1) - Production-ready Buzz Relay. Everything included. Zero config. Just works.
- [OpenHands | Open Source Devin Alternative](https://railway.com/deploy/openhands-ai-agent) - Self-host the OpenHands AI software engineer agent on Railway
- [Hermes Agent All-in-One | WebUI + Admin Panel + Gateway - No Terminal Setup](https://railway.com/deploy/hermes-all-in-one-or-webui-admin-panel-g) - Unified agent across apps, browser setup, auto-ready, skill, and automation
- [newapi](https://railway.com/deploy/newapi) - new-api (AI) railway template
- [Onyx](https://railway.com/deploy/onyx) - Onyx - Open Source AI Platform - AI Chat with advanced features
- [OpenCode for Railway](https://railway.com/deploy/opencode-for-railway) - Pre-configured OpenCode server template for Railway.
- [OpenClaw (All-in-One Bundle)](https://railway.com/deploy/openclaw-all-in-one-bundle) - OpenClaw + Web Wizard + Lite Panel + Web TUI + Browser + Searxng Search
- [Claude Code SSH](https://railway.com/deploy/claude-code-mobile) - Run Claude Code safely from anywhere via SSH in 2 simple steps.
- [Hermes Agent Template](https://railway.com/deploy/hermes-agent-template) - Deploy Hermes Agent as a secure web dashboard with persistent storage.
- [Hermes Agent - Custom Dashboard](https://railway.com/deploy/hermes-agent-2) - Self-Hosted Hermes AI Agent with Telegram, Discord & Slack
- [Ubuntu Linux + Claude Code (Browser Terminal + SSH)](https://railway.com/deploy/claude-code-box) - Claude Code, Codex and Gemini CLI on Ubuntu 24.04, home dir on a volume
- [9router [Updated Sep '26]](https://railway.com/deploy/9router-1) - 9Router [Sep '26] (Self-Hosted AI Gateway & LLM Router) Self Host
- [Qdrant](https://railway.com/deploy/i1tz3T) - Open-source vector database for AI, RAG and semantic search
- [Paperclip (Official Image)](https://railway.com/deploy/paperclip-official-1) - Run AI agent teams (Claude Code, Codex, Gemini) with budgets and approvals
- [Kokoro TTS Lite | OpenAI-Compatible Voice API, Runs on the Free Plan](https://railway.com/deploy/kokoro-tts-lite) - OpenAI-compatible TTS on ONNX. Fits 512MB, runs on the Free plan.
- [Hermes Agent by Nous Research](https://railway.com/deploy/hermes-agent-railway-template) - Official self-improving Hermes Agent — grows and learns with every use
- [flowise-railway](https://railway.com/deploy/pn4G8S) - Flowise - low-code LLM apps builder
- [Ubuntu SSH + Claude Code](https://railway.com/deploy/ubuntu-ssh-claude-workstation) - SSH-in Ubuntu workstation with Claude Code
- [Firecrawl | Open-Source Web Scraper - Apify, Scrapy Alternative](https://railway.com/deploy/firecrawl-web-scraping-api) - Self-host Firecrawl on Railway. Scrape/Crawl website to .md, LLM extraction
- [LiteLLM](https://railway.com/deploy/litellm) - Unified LLM gateway — route to 100+ providers through one OpenAI-compatible
- [OpenHands](https://railway.com/deploy/openhands) - openhands: AI agent for coding and deployment. open hands
- [Langflow](https://railway.com/deploy/LcJLmZ) - Drag-and-drop interface for building AI-powered agents and workflows.
- [Codex CLI - Terminal and Browser](https://railway.com/deploy/codex-cli-terminal-and-browser) - Run the OpenAI Codex CLI on a persistent Railway box, TUI or browser
- [Agent Zero](https://railway.com/deploy/agent-zero) - Self-hosted general-purpose AI agent with terminal & code exec
- [hermes-agent-railway](https://railway.com/deploy/hermes-agent-railw-1) - Deploy a secure, persistent Hermes Agent from the official image
- [Open WebUI](https://railway.com/deploy/open-web-ui) - Private multi-model AI chat [Oct'26] — OpenAI, Claude, Ollama, built-in RAG
- [manifest](https://railway.com/deploy/wild-wild) - Plug your AI agents into any provider.
- [Ollama — Private LLM & Embeddings API on Railway](https://railway.com/deploy/ollama-llm-api) - Self-host Ollama: private LLM & embeddings API, OpenAI-compatible. No GPU.
- [OpenClaw](https://railway.com/deploy/openclaw-private-agent) - OpenClaw [Oct'26]: private AI agent for ~$1/month. No ChatGPT subscription
- [Alquimia Stack](https://railway.com/deploy/alquimia-stack-2) - Deploy Alquimia Stack on Railway
- [OpenWebUI [Updated Sep ’26]](https://railway.com/deploy/open-webui) - OpenWebUI [Sep ’26] (Chat with Local or Remote AI Models) Self Host
- [Crawl4AI](https://railway.com/deploy/Owzyy8) - Crawl4AI is an AI-ready web crawler for LLMs, AI agents, & data pipelines.
- [9router](https://railway.com/deploy/9router-4) - Self-hosted AI router for Cursor, Claude, Gemini, OpenAI & 60+ AI providers
- [Lovense Connector for ChatGPT](https://railway.com/deploy/lovense-connector-for-chatgpt) - Private Lovense MCP for ChatGPT with OAuth and multi-device control.
- [Ollama [Updated Jun '26]](https://railway.com/deploy/ollama-or-self-host-open-source-llms) - [Jun '26] Self-host your own LLM model server with a single click.
- [Sim Studio](https://railway.com/deploy/sim-studio) - A lightweight, user-friendly platform for building AI agent workflows.
- [🪰 Fruit Fly Brain](https://railway.com/deploy/fruit-fly-brain) - Deploy and host your own 🪰 Fruit Fly Brain on Railway!
- [OpenClaw](https://railway.com/deploy/openclaw-ai) - The personal AI assistant that really does things.
- [deputies-monolith](https://railway.com/deploy/deputies-monolith) - Deputies, a control plane for delegated engineering agents.
- [Kokoro TTS API | (Just Updated) OpenAI-Compatible Voice, Hobby Plan Recommended](https://railway.com/deploy/kokoro-tts-api) - OpenAI-compatible TTS API. Needs 1.5GB RAM, Hobby plan or above.
- [Open Source LLM Models](https://railway.com/deploy/open-source-llm-models) - [Jun '26] Self host open LLMs + connect your ChatGPT Account
- [Hermes Agent with Authenticator 2FA](https://railway.com/deploy/hermes-agent-with-authenticator-2fa) - Hermes Agent with persistent storage and password + authenticator 2FA.
- [Toast MCP](https://railway.com/deploy/toast-mcp) - Connect Claude to your Toast POS. Read-only, your credentials, 55 tools.
- [Hermes Kanban Web](https://railway.com/deploy/hermes-kanban-web-1) - Self-hosted Kanban board powered by the Hermes Agent with web terminal
- [Hermes Agent by Nous Research](https://railway.com/deploy/hermes-official-dashboard) - A self-hosted Hermes Agent using Nous Research's official Docker image
- [pi](https://railway.com/deploy/pi) - Run the Pi AI coding agent on Railway — browser terminal or SSH
- [Ollama](https://railway.com/deploy/ollama-stack-llm-web-ui) - 1-click deploy Ollama LLM with Open WebUI. Ready to use 🤖
- [Browser Use](https://railway.com/deploy/browser-use) - AI browser automation runtime with Chromium. Deploy in 1-click 🚀
- [Buzz (Self-Hosted)](https://railway.com/deploy/buzz-self-hosted) - Self-host Buzz with PostgreSQL, Redis, S3 and secure mobile pairing.
- [Langflow](https://railway.com/deploy/langflow-ai) - Low-code app builder for RAG and multi-agent AI applications.
- [Mem0 | The Memory Layer for Personalized AI](https://railway.com/deploy/mem0) - Self Host Mem0. Give your AI agents persistent memory
- [LiveKit - Voice Agent](https://railway.com/deploy/livekit) - Deploy and Host LiveKit with Railway
- [Tinykit](https://railway.com/deploy/tinykit) - Self-hosted Lovable/v0 alternative. Real-time database & storage included.
- [OpenAI-Proxy](https://railway.com/deploy/HLP0Ub) - OpenAI Proxy (100+ LLMs) - OpenAI, Azure, Bedrock, Anthropic, HuggingFace
- [CLIProxyAPI | (Just Updated) Your Subscription as an API, Key-Locked From Boot](https://railway.com/deploy/cliproxyapi) - OpenAI-compatible proxy for your CLI plans, API key enforced from boot
- [Hermes Agent (Slim) — Cheapest Self-Improving AI Agent](https://railway.com/deploy/hermes-agent-slim-cheapest-self-improvin) - Memory-efficient self-hosted Hermes AI agent, self-improving, 24/7
- [Agentic Data Stack Lite (Langfuse Cloud)](https://railway.com/deploy/agentic-data-stack-lite-langfuse-cloud) - The Agentic Data Stack instrumented with Langfuse Cloud
- [Speaches | Open-Source OpenAI Whisper Alternative on Railway](https://railway.com/deploy/speaches) - Self Host Speaches — OpenAI-compatible STT/TTS API server
- [hindsight](https://railway.com/deploy/hindsight) - Long-term AI memory system with API+Control Plane in one deploy
- [AutoGen Studio](https://railway.com/deploy/autogen-studio) - Build Multi-Agent Apps and rapidly prototype AI agents
- [Qwen3 Chat](https://railway.com/deploy/qwen-chat) - [Jun '26] Self-host Alibaba's Qwen models locally, using Ollama+OpenWebUI.
- [SillyTavern](https://railway.com/deploy/sillytavern-official-image) - SillyTavern on the official image, password-protected, chats on a volume
- [DeepSeek Harness + Web UI](https://railway.com/deploy/deepseek-harness-w-1) - Secure, persistent DeepSeek Harness Web UI with protected browser access.
- [Bolt.diy](https://railway.com/deploy/bolt-diy) - Chat with an AI to build and run web apps in your browser
- [Buzz](https://railway.com/deploy/buzz-relay-agent-workspace) - Self-hosted Slack + GitHub alternative for humans & AI agents [Oct'26]
- [SearXNG Search API — Private Metasearch for AI](https://railway.com/deploy/searxng-valkey-ai-search) - Self-hosted search API for AI agents — JSON, no API keys, Valkey
- [Headroom](https://railway.com/deploy/headroom) - Headroom compression proxy. Cuts LLM token usage
- [Open WebUI Pro Stack One-click Deploy](https://railway.com/deploy/open-webui-pro-stack) - AI stack: open webui, litellm, pg & redis. openwebui open web ui openweb ui
- [Multica](https://railway.com/deploy/multica) - Platform for building and orchestrating AI agents.
- [MCP Servers](https://railway.com/deploy/mcp-servers) - MCP gateway — expose web fetch, memory, and reasoning tools to AI clients.
- [Jupyter Notebook & JupyterLab Cloud Deployment (with Spark Support)](https://railway.com/deploy/jupyter-notebook-and-jupyterlab-cloud-de) - [Jun '26] One-click JupyterLab hosting for big data, and ML workflows.
- [discord-ai](https://railway.com/deploy/discord-ai) - Discord AI bot using OPENROUTER APIs
- [N8N using PostgreSQL](https://railway.com/deploy/vFXgsT) - n8n with PostgreSQL, auto-updates, and Community Nodes tool usage support
- [Open WebUI](https://railway.com/deploy/2yb5lj) - User-friendly ChatGPT UI alternative designed to operate offline. 
- [OpenClaw — Secure One-Click Setup](https://railway.com/deploy/openclaw-secure-on-1) - Secure OpenClaw with browser setup, persistent data, and one-click deploy.
- [Open WebUI](https://railway.com/deploy/open-webui-ai-platform) - Your self-hosted interface for local and cloud AI models.
- [SillyTavern | (Just Updated) AI Character Chat, No Stranger Can Claim Your Admin](https://railway.com/deploy/sillytavern-or-just-updated-ai-character) - SillyTavern AI chat frontend, admin password-protected from first boot
- [Qwen3 Chat](https://railway.com/deploy/qwen3-ollama-openwebui) - Run Qwen3 models privately — Ollama + chat UI, or via API
- [Turn ChatGPT into OpenAI API - Chat2API](https://railway.com/deploy/chat2api) - Stream responses, upload files & rotate multiple accounts.
- [Universal API Gateway - New API](https://railway.com/deploy/new-api-or-universal-ai-gateway) - Deploy this new API gateway for OpenAI, Gemini, and Claude.
- [AdClaw](https://railway.com/deploy/adclaw) - AI marketing agent team - 130+ skills, multi-agent personas, multi-channel 
- [Paperclip](https://railway.com/deploy/paperclip-2) - Run an AI company with autonomous agents
- [hermes-webui](https://railway.com/deploy/hermes-webui) - One-click Hermes Web UI: secure access, persistent state, zero ops
- [Lightpanda Headless Browser](https://railway.com/deploy/lightpanda-headless-browser) - Ultra-lightweight headless browser for AI agents and web automation.
- [DeepSeek](https://railway.com/deploy/deepseek) - [Jun '26] Self Host DeepSeek models and chat via a chat UI
- [Ollama API](https://railway.com/deploy/ollama-api) - A powerful platform for running AI models securely through API
- [FlowiseAI](https://railway.com/deploy/A7Dwg9) - Flowise AI with persisted volume, PostGIS & private networking.
- [Kokoro-TTS](https://railway.com/deploy/kokoro-tts) - Turns written text into natural-sounding speech, in 60+ voices
- [Ollama AI](https://railway.com/deploy/ollama-ai) - Run your private AI models with a clean web interface, ready in 1 click.
- [T3 Code](https://railway.com/deploy/t3-code) - T3 code hosted on railway with the web app open to the public
- [Hermes Agent (Full) — Self-Improving AI Agent with Official Dashboard](https://railway.com/deploy/hermes-agent-full-self-improving-ai-agen) - Self-hosted Hermes AI agent + official Nous dashboard, self-improving
- [otari](https://railway.com/deploy/otari) - OpenAI-compatible LLM gateway: virtual keys, budgets, usage tracking
- [Flowise](https://railway.com/deploy/flowise-ai) - [Jul'26] Flowise + persistent storage for visual AI agents & LLM workflows.
- [Bifrost](https://railway.com/deploy/bifrost-1) - Fastest enterprise AI gateway (50x faster than LiteLLM).
- [DeerFlow V2](https://railway.com/deploy/deerflow) - AI powered Deep Research - Open Source, by ByteDance ai
- [Bolt DIY](https://railway.com/deploy/bolt) - [Updated Sep '26] Deploy and Host Bolt DIY with Railway
- [neo4j](https://railway.com/deploy/ZVljtU) - Creates a Neo4j graph database and volume for persistent storage.
- [Dify LLM App Platform + MinIO](https://railway.com/deploy/dify-llm-app-platform-minio) - Dify LLM app platform with pgvector, plugin daemon, sandbox and MinIO
- [Vault Cortex](https://railway.com/deploy/vault-cortex) - MCP server for your Obsidian vault: search, memory, tasks, files, OAuth 2.1
- [Ollama](https://railway.com/deploy/ollama-railway) - Open-source local LLM runtime: chat, embeddings, OpenAI-compatible /v1 API
- [Langfuse v4.52.0 (Production ready AI Observability)](https://railway.com/deploy/langfuse-v410-production-ready-ai-observ) - Open-source LLM platform for prompt management, tracing, evals & metrics
- [N8n Self Hosted AI Stack](https://railway.com/deploy/n8n-self-hosted-ai-stack) - A fully private, self hosted automation and AI stack.
- [Agentmemory - Self-hosted Memory for AI Agents](https://railway.com/deploy/agentmemory-self-hosted-memory-for-ai-ag) - Persistent memory server for MCP agents. agentmemory agent memory
- [TradingAgents By Tauric Research | Open Source Multi-Agent Stock Analysis](https://railway.com/deploy/trading-agents) - Self-host TradingAgents. Multi-agent stock analysis w/ OpenAI/Claude/Gemini
- [hermes agent: easy self-improving option for open claw, claude code, codex and more](https://railway.com/deploy/hermes-agent-easy-self-improving-option-) - The simplest curated setup for an Hermes Agent. Curated by impacte.tech
- [Digital Twin AI](https://railway.com/deploy/digital-twin-ai) - Browser-based VS Code with Claude Code CLI pre-installed
- [AI-Trader](https://railway.com/deploy/ai-trader) - AI-Trader with MCP tools, UI dashboard. AI Trader IA Trader
- [AnythingLLM [Updated Sep ’26]](https://railway.com/deploy/anything-llm) - AnythingLLM [Sep ’26] (Chat, RAG & Local AI Assistants) Self Host
- [MLflow](https://railway.com/deploy/mlflow) - Deploy and Host a production-ready MLflow instance on Railway
- [Docling - OCR anything](https://railway.com/deploy/docling-ocr-anything) - Get any document AI / LLM Ready
- [N8N (w/ pgvector)](https://railway.com/deploy/msyuqd) - N8N AI complete kit. N8N with Postgres pgvector enabled.
- [Bolt.diy NEW](https://railway.com/deploy/abVHie) - official open source version of Bolt.new
- [LibreChat - Lite](https://railway.com/deploy/_fTxzh) - lite version LibreChat without RAG
- [arize-phoenix](https://railway.com/deploy/PTHRoq) - AI Observability & Evaluation
- [flowise-with-n8n-ai-automation](https://railway.com/deploy/Xx4_lu) - n8n ai automation template
- [NextChat](https://railway.com/deploy/Mvjjpn) - An open-source ChatGPT UI alternative.
- [9Router AI Gateway [Sep'26]](https://railway.com/deploy/9router-ai-gateway-aug26) - Self-hosted AI gateway for Claude Code, Cursor & 40+ LLM providers
- [LibreChat](https://railway.com/deploy/libre-chat) - Private multi-provider AI chat: agents, code interpreter, document RAG
- [Mirofish](https://railway.com/deploy/mirofish) - A Simple and Universal Swarm Intelligence Engine, Predicting Anything.
- [Jupyter Notebook](https://railway.com/deploy/jupyter-notebook) - Deploy Jupyter Lab | Jupyter Notebook on Railway
- [Ollama + Open WebUI | Fast CPU Inference, Model Preloaded](https://railway.com/deploy/ollama-open-webui-or-fast-cpu-inference-) - Ollama tuned for Railway CPU: ~78x faster, model pulled on first boot.
- [Dify Production](https://railway.com/deploy/dify-production) - Open-source LLM app platform — workflows, agents, RAG. 10-service stack
- [odysseus](https://railway.com/deploy/odysseus-1) - Self-hosted AI workspace with chat, agents, memory, and research
- [Open WebUI](https://railway.com/deploy/open-webui-standard) - Run AI on your terms. Connect any model, extend code, protect data
- [OpenHands Coding Agent](https://railway.com/deploy/openhands-coding-agent) - Self-hosted autonomous AI engineer. Open-source, model-agnostic, free
- [Dify | Open-Source AI/LLM App Builder](https://railway.com/deploy/dify-ai-workflow) - Self-host Dify on Railway — open-source LLM app builder with RAG and agents
- [Label Studio — Open Source Alternative to Scale AI & Labelbox](https://railway.com/deploy/self-host-label-studio) - Self Host Label Studio. Data labeling for images, text, audio & video.
- [Kokoro Web](https://railway.com/deploy/kokoro-web) - Free AI Voice Generator with API integration
- [Bolt DIY [Updated Sep ’26]](https://railway.com/deploy/boltdiy) - Bolt DIY [Sep ’26] (AI App/Web Builder, Lovable & V0 alternative) Self Host
- [Langflow [Updated Sep '26]](https://railway.com/deploy/langflow) - Langflow [Sep '26] (LLM Workflow Builder, Flowise alternative) Self Host
- [OpenWebUI](https://railway.com/deploy/openwebui-team-production) - Self-host Open WebUI [Oct'26]— durable Postgres, scalable sessions
- [DSH + LongMemory](https://railway.com/deploy/dsh-longmemory) - DSH, DeepSeek's coding agent, with LongMemory as its MCP memory server
- [llama.cpp + Open WebUI](https://railway.com/deploy/llamacpp-open-webui) - Deploys llama.cpp (free tokens) +Open WebUI
- [Claude Code | Cloud Dev Box (Web Terminal + SSH)](https://railway.com/deploy/claude-code) - Claude Code, git, gh, Node, Python. Browser terminal + SSH, home persists.
- [Rakazo](https://railway.com/deploy/rakazo) - Persistent AI teammates with memory, routines, and their own computers.
- [OpenShorts](https://railway.com/deploy/openshorts-1) - AI short generator from your long form video
- [Dify](https://railway.com/deploy/dify-ai) - Visual builder for AI apps, chatbots and document workflows
- [Docling](https://railway.com/deploy/docling) - LlamaParse Alternative. Convert PDF, DOCX and scans to Markdown and JSON
- [Firecrawl](https://railway.com/deploy/firecrawl-railway) - Turn any website into LLM-ready Markdown with self-hosted Firecrawl
- [PDFMathTranslate](https://railway.com/deploy/pdfmathtranslate) - Translate academic PDFs keeping layout, formulas and charts intact
- [Sandbox](https://railway.com/deploy/sandbox-1) - All-in-one AI agent sandbox with browser, shell, MCP, and VSCode.
- [Marqo](https://railway.com/deploy/marqo) - Deploy and Host Marqo + Navigator Ecommerce Search and Discovery on Railway
- [GBrain Agentic Memory](https://railway.com/deploy/gbrain-agentic-memory) - Get your own self hosted gbrain agentic memory in 2 simple steps
- [lacuna](https://railway.com/deploy/lacuna) - Turn scanned PDFs into a searchable archive and reader site, with OCR
- [OmniRoute — LLM Gateway with Prompt Compression](https://railway.com/deploy/omniroute-llm-gateway-compression) - Self-host OmniRoute — cut token costs, unify LLM providers
- [Langflow](https://railway.com/deploy/langflow-1) - Visual builder for AI agents, workflows, and API applications.
- [Image to Video AI Railway Template](https://railway.com/deploy/image-to-video-ai-railway-template) - Deploy a minimal Image to Video AI starter page on Railway.
- [Supermemory](https://railway.com/deploy/supermemory) - Self-hosted memory and context engine for AI
- [stella](https://railway.com/deploy/stella) - Self-host stella with web, API, Postgres, Redis, and object storage.
- [hermes-studio](https://railway.com/deploy/hermes-studio) - Self-hosted Hermes Agent dashboard with persistent state
- [Paperclip — AI Agent Company OS on Railway](https://railway.com/deploy/paperclip-ai-agent) - Self-host Paperclip: org charts, goals & budgets for your AI agent team.
- [dograh](https://railway.com/deploy/dograh) - Open Source Voice Agent Platform
- [OpenHuman](https://railway.com/deploy/openhuman) - OpenHuman core with persistent memory and secure RPC. openhuman Open Human
- [MinerU | Document Parsing API](https://railway.com/deploy/mineru) - Self-host MinerU PDF parser with an OpenAI-compatible API
- [Perplexica | Open Source Perplexity AI Alternative](https://railway.com/deploy/perplexica-ai-search) - Self-host Perplexica AI. AI-powered search engine with citations.
- [Letta | Open-Source Stateful AI Agent Framework](https://railway.com/deploy/letta-ai-agent) - Self-host Letta (MemGPT). Build AI agents that remember everything
- [booa-hermes-template](https://railway.com/deploy/booa-hermes-template) - Your BOOA as a self-hosted AI agent.
- [nanobot](https://railway.com/deploy/nanobot-4) - [Jul'26] Minimal OpenClaw lightweight AI agent. It works, 1-click deploy🚀
- [OpenClaw Railway Template](https://railway.com/deploy/openclaw-railway-t-1) - [Jun '26] Simple 1-click deploy OpenClaw (Clawdbot)  on VPS | TUI Support
- [Agenta IA](https://railway.com/deploy/agenta-railway-template) - Agenta (open-source LLMOps) on Railway - 14-service OSS stack (1 click)
- [Paperclip (Zero Config)](https://railway.com/deploy/paperclip-zero-config) - Run a company of AI agents. Secure login, Postgres, zero config.
- [Laya | Self-Hosted Jev-Compatible Decision API](https://railway.com/deploy/laya) - Open-weights Laya decision model behind a TypeSafe/Jev-compatible API
- [Skyvern Browser Agent + Auth Gateway](https://railway.com/deploy/skyvern-browser-agent-auth-gateway) - Skyvern browser agent with a password protected UI gateway
- [Hermes Agent](https://railway.com/deploy/kit-hermes-build-0920) - Hermes Agent: self-hosted AI agent for Telegram, Discord, Slack with memory
- [Mem0 Memory API + pgvector](https://railway.com/deploy/mem0-memory-api-pgvector) - Mem0 memory layer for AI agents: REST API, dashboard and pgvector Postgres
- [Graphiti MCP + Neo4j](https://railway.com/deploy/graphiti-mcp-neo4j) - Knowledge-graph memory MCP server for AI agents, with persistent Neo4j
- [Vane (Perplexica) + SearXNG](https://railway.com/deploy/vane-perplexica-searxng) - Perplexity-style AI answering engine with a private SearXNG backend
- [JupyterLab](https://railway.com/deploy/jupyter) - JupyterLab notebooks, terminal and file browser in one hosted workspace
- [Mastra](https://railway.com/deploy/mastra) - A TypeScript AI agent platform with Studio, workflows & persistent memory.
- [Astrbot](https://railway.com/deploy/astrbot-1) - AI Agent Assistant & Chatbot Platform - multi-platform, LLM-agnostic
- [Buzz Relay](https://railway.com/deploy/buzz-relay-2) - Nostr relay with built-in Git hosting. Auto-wired Postgres, Redis, MinIO.
- [Langflow](https://railway.com/deploy/langflow-agent-builder) - Visual canvas for building AI agents and LLM workflows
- [Nautilo](https://railway.com/deploy/nautilo) - Run a private workspace for people and AI agents.
- [Ombre-Brain](https://railway.com/deploy/ombre-brain) - Self-hosted emotional memory for AI with Obsidian storage and search.
- [Multica (Prebuilt)](https://railway.com/deploy/multica-prebuilt) - AI coding agent issue tracker for Claude Code, Codex, Cursor and 17 more
- [Google Drive CRUD](https://railway.com/deploy/google-drive-crud) - Your own private Google Drive, Docs, Sheets and Slides connector for Claude
- [ZeroClaw — Lightweight Self-Hosted AI Agent](https://railway.com/deploy/zeroclaw-ai-agent) - Self-host ZeroClaw — a tiny Rust AI agent for Discord & more
- [Buzz by Block](https://railway.com/deploy/buzz-railway-template) - A workspace where humans and agents build together, on a relay you own.
- [Paperclip Latest Version [Aug 26]](https://railway.com/deploy/paperclip-latest-version) - Deploy Latest Version of Paperclip in one click.
- [Coshell](https://railway.com/deploy/coshell) - Multiplayer AI coding workspace where your team and AI agents work together
- [GPT Load](https://railway.com/deploy/gptload) - Distribute requests and load balance multiple GPT API keys
- [Llama 3.2 1B — Lightweight Private AI Chat](https://railway.com/deploy/llama-3-2-1b-private-chat) - Self-host Llama 3.2 1B — fast, lightweight private chat on CPU
- [Paperclip](https://railway.com/deploy/paperclip-3) - Host Paperclip with authenticated access and persistent agent state.
- [Buzz](https://railway.com/deploy/buzz-inloop-studio) - Self-host Buzz Relay 0.2.0 with private, persistent Railway services
- [AstrBot - OpenClaw alternative](https://railway.com/deploy/astrbot-openclaw-alternative) - OpenClaw alternative: AI agent framework with IM & LLM integration
- [Mastra Server With Studio](https://railway.com/deploy/mastra-server-with-studio) - Mastra server with Studio UI - build, test, and manage AI agents
- [Hermes Agent deployer by aiexperts.com](https://railway.com/deploy/aiexperts-hermes-agent) - Hermes Agent on Railway - Preconfigured, more secure, and faster deployment
- [OpenClaw with Ollama](https://railway.com/deploy/openclaw-ollama) - Launch OpenClaw in minutes with open models powered by Ollama
- [RAGFlow | Open Source RAG Engine](https://railway.com/deploy/ragflow-rag-engine) - Self-host RAGFlow. Chat with your PDFs, contracts, papers & more
- [vLLM | High-Throughput LLM Serving](https://railway.com/deploy/vllm) - Self-host vLLM on Railway with an OpenAI-compatible API
- [hermes-workspace](https://railway.com/deploy/hermes-workspace) - AI command center for Hermes: chat, files, memory,  skills, terminal
- [Bifrost AI Gateway](https://railway.com/deploy/bifrost-ai-gateway) - Go from zero to production-ready AI gateway in under a minute.
- [InvokeAI](https://railway.com/deploy/invokeai) - A powerful creative engine and WebUI for Stable Diffusion image generation
- [Flowise | Build AI Agents](https://railway.com/deploy/flowise-or-build-a-2) - [Jun '26] Low-code developer tool for building LLM Applications
- [Dify v1.17.1](https://railway.com/deploy/TiT2V6) - Open-source AI platform with BaaS & LLMOps for all users.
- [Cognee Graph Memory API + pgvector](https://railway.com/deploy/cognee-graph-memory-api-pgvector) - Cognee graph memory API for agents, pgvector Postgres and auth gateway
- [Paperclip](https://railway.com/deploy/paperclip-official-image) - Run a team of AI agents: Claude Code, Codex and more, with Postgres
- [Hindsight Memory Server](https://railway.com/deploy/hindsight-agent-memory-secure-slim) - Long-term memory for AI agents over MCP and REST, with auth and pgvector
- [Orca ADE](https://railway.com/deploy/orca-ade) - Orca ADE server: Claude Code, Codex, Pi in parallel worktrees, any device
- [9Router](https://railway.com/deploy/9router-3) - 9Router AI gateway: one OpenAI-compatible API for 40+ LLM providers
- [LobbyStack](https://railway.com/deploy/lobbystack) - Open-source AI receptionist that answers calls and books appointments.
- [LobeHub Server + Postgres](https://railway.com/deploy/lobehub-server-postgres) - LobeHub 2 server mode: AI chat and agents with Postgres, Redis and S3
- [Multica AI](https://railway.com/deploy/multica-ai) - Manage AI coding agents like teammates. Assign work, track progress & ship.
- [Agentlings](https://railway.com/deploy/agentlings) - Your own horde of agentlings, working real jobs you review
- [openchamber-devbox](https://railway.com/deploy/openchamber-devbox) - Self-hosted AI developer workstation with OpenChamber & OpenCode
- [Hindsight](https://railway.com/deploy/hindsight-memory-api) - Open-source long-term memory.Retain/Recall/Reflect memory API
- [LiteLLM Proxy](https://railway.com/deploy/litellm-proxy-1) - LLM proxy for Claude, GPT & Gemini with keys, budgets & caching
- [DeepSeek Harness on NixOS | (Just Updated) Agent With All of Nixpkgs](https://railway.com/deploy/deepseek-harness-on-nixos-or-just-update) - DeepSeek's coding agent on nix. Password-gated, any package it needs.
- [Codex Box | (Just Updated) Remote Dev Machine Logged In, Persistent](https://railway.com/deploy/codex-box-v0147-or-remote-dev-machine-pi) - Remote dev machine that stays logged in, with the workspace persisted
- [lobehub](https://railway.com/deploy/lobehub) - Deploy and Host Lobehub. Lobechat lobe hub lobe chat
- [DeepSeek Chat — Ollama + Open WebUI, Private AI](https://railway.com/deploy/deepseek-ollama-openwebui) - Run distilled DeepSeek models privately — Ollama + chat UI
- [Flowise](https://railway.com/deploy/flowise-4) - Self-hosted Flowise: build AI agents & LLM workflows visually. Your keys.
- [Comfy UI](https://railway.com/deploy/comfy-ui) - Self-hosted node-based UI for Stable Diffusion image generation.
- [LLaMA.cpp](https://railway.com/deploy/llamacpp-1) - llama.cpp — OpenAI-compatible LLM inference server, bring your own GGUF
- [qwen3-tts](https://railway.com/deploy/qwen3-tts) - Self-hosted TTS API, OpenAI-compatible. CPU-only Qwen3-TTS - no GPU bill.
- [Blinko](https://railway.com/deploy/blinko) - Blinko — AI-powered note-taking with RAG chat over your notes
- [OpenClaw (Clawdbot / Moltbot) [Aug'26]](https://railway.com/deploy/openclaw-jun26) - Self-hosted AI assistant (Clawdbot) for Telegram, WhatsApp, Slack, Discord
- [Paperclip - AI Agent Company](https://railway.com/deploy/paperclip-ai-agent-company) - Open-source control plane for AI agents. Includes managed Postgres
- [hermes-all-in-one](https://railway.com/deploy/hermes-all-in-one) - Hermes Agent with WebUI and Control Plane
- [luthien-proxy-oneclick](https://railway.com/deploy/luthien-proxy-oneclick) - Deploy and Host luthien-proxy-oneclick with Railway
- [Hermes Agent](https://railway.com/deploy/hermes-agent-1) - Nous Research’s self-improving AI agent that learns across sessions.
- [AgentField Engineering Team](https://railway.com/deploy/agentfield-engineering-team) - AI Powered Engineering Team
- [Faster Whisper](https://railway.com/deploy/faster-whisper) - Private Speech-to-Text STT & Text-to-Speech TTS, OpenAI & 52 voices
- [PicoClaw](https://railway.com/deploy/picoclaw) - Ultra-Efficient AI Assistant in Go
- [SearXNG API](https://railway.com/deploy/searxng-api) - A pre-configured SearXNG optimized for private and LLM tool use.
- [open-wearables](https://railway.com/deploy/open-wearables) - Unify wearable health data through one AI-ready API.
- [Llama in Cloud](https://railway.com/deploy/llama-in-cloud) - [Jun '26] Host Meta's Llama models privately, using Ollama + OpenWebUI.
- [llama-3.2-1b](https://railway.com/deploy/llama-32-1b) - Deploy and host llama 3.2-1b on Railway.
- [Crawl4AI](https://railway.com/deploy/crawl4ai) - Open-source LLM-friendly web crawler with REST API and MCP integration.
- [Portkey AI - Gateway](https://railway.com/deploy/jOsex-) - A Blazing Fast AI Gateway. Route to 100+ LLMs with 1 fast & friendly API.
- [Langfuse v2](https://railway.com/deploy/gmbqa_) - Open source observability and analytics for LLM applications
- [Chat Chat](https://railway.com/deploy/-WWW5r) - Chat Chat, your own unified chat and search to AI platform.
- [Multica Self-Hosted](https://railway.com/deploy/multica-self-hosted) - Humans and AI coding agents on one task board. Sign-up limited to you.
- [OpenCode | AI Coding Agent in Your Browser, Free Models Included](https://railway.com/deploy/opencode-cloud) - Open-source AI coding agent in your browser. Free models, SSH, saved home.
- [Whoop MCP Server](https://railway.com/deploy/whoop-mcp-server) - Ask Claude, ChatGPT or another MCP app about your WHOOP data.
- [Honcho](https://railway.com/deploy/honcho-1) - Self-hosted memory server for AI agents (unofficial Honcho stack)
- [OpenDesign](https://railway.com/deploy/opendesign) - Open-source Claude Design alternative, with Claude Code, Codex and OpenCode
- [Orca ADE (w/ Relay)](https://railway.com/deploy/orca-ade-w-relay) - Orca agent server + your own relay: Claude Code, Codex, Pi from any device
- [Smart Marking](https://railway.com/deploy/smart-marking-1) - AI marking for scripts and Computing files, teacher-checked before release
- [orca](https://railway.com/deploy/orca) - Remote Orca server: an agent development environment you pair devices to
- [Tallylamp](https://railway.com/deploy/tallylamp) - MCP browsers with saved profiles, human takeover, guest links and proxies.
- [OpenBot (Grok Bot Alternative)](https://railway.com/deploy/openbot) - OpenBot: open-source AI coworkers with their own browser, files and tools
- [LocalAI](https://railway.com/deploy/local-ai) - Run open-source AI models on your own server through one API
- [Personal AI Assistant for Telegram](https://railway.com/deploy/personal-ai-assistant) - Private Telegram AI with memory, web search, voice, vision & files.
- [claude-stat-remote](https://railway.com/deploy/claude-stat-remote) - Monitor sesi Claude Code real-time di Android, di mana saja lewat Railway
- [Browser Use](https://railway.com/deploy/browser-use-railway-template) - AI browser automation runtime with Chromium. Deploy in 1-click 🚀
- [OpenClaw Office](https://railway.com/deploy/openclaw-office) - A virtual office for OpenClaw agents to collaborate, execute, & work live.
- [Open Notebook](https://railway.com/deploy/opennotebook) - AI research notebooks that chat with your documents and files.
- [TrueForge](https://railway.com/deploy/trueforge) - Self-host TrueForge with protected UI, PostgreSQL and Redis.
- [LiteLLM](https://railway.com/deploy/litellm-gateway) - One API for every AI model, with keys, budgets and usage
- [LabelStudio](https://railway.com/deploy/label-studio-1) - Tool for labeling data to train machine learning models
- [Pocket TTS](https://railway.com/deploy/pocket-tts-1) - CPU-only text-to-speech API. Send text, get audio back.
- [Follow The Thread](https://railway.com/deploy/follow-the-thread) - AI Powered Multi Threaded Research Agent
- [PipesHub](https://railway.com/deploy/pipeshub) - AI context layer with enterprise search, connectors, agents, and RAG
- [OpenClaw — Personal AI Assistant](https://railway.com/deploy/openclaw-ai-assistant) - Let your AI assistant handle the work, ready in just 1 click.
- [Buzz Relay](https://railway.com/deploy/buzz-2) - Your own Buzz Relay for humans and agents. Zero config. Just works.
- [ChatGPT Web Midjourney Proxy](https://railway.com/deploy/chatgpt-web-midjourney-proxy) - A web UI supporting AI chat, art, music, video, dance, and live voice.
- [Buzz Relay | Block AI Agent Workspace, Repos Survive Redeploy](https://railway.com/deploy/buzz-relay-or-block-ai-agent-workspace-r) - Block's Buzz relay: chat, code review and git for humans and AI agents
- [Open WebUI and Ollama](https://railway.com/deploy/ollama-open-webui) - Private AI: Ollama inference behind an authenticated Open WebUI chat UI
- [Tinykit](https://railway.com/deploy/tinykit-ai-app-builder) - Self-host an AI app builder — DB, storage & 100+ apps, one service
- [GProxy](https://railway.com/deploy/gproxy) - Multi-provider LLM proxy with built-in admin frontend
- [Relaticle agent-native CRM](https://railway.com/deploy/relaticle-agent-native-crm) - Agent-native CRM with durable records, uploads, PostgreSQL, and Redis.
- [VoAPI](https://railway.com/deploy/voapi) - Advanced AI API proxy and gateway for model access distribution.
- [gcli2api](https://railway.com/deploy/gcli2api) - Convert GeminiCLI and Antigravity to OpenAI,GEMINI Claude API interfaces
- [Company Brain Gateway](https://railway.com/deploy/company-brain-gateway) - One isolated private GitHub brain for Codex/Claude in buyer-owned Railway.
- [Flowise - Railway Template](https://railway.com/deploy/flowise-railway-template) - Flowise with web UI, auth, and persistent storage. It just works 🤖
- [WebUI](https://railway.com/deploy/webui-1) - Deploy OpenWebUI with Postgres & Redis
- [AnythingLLM — Private RAG & AI Agent Workspace](https://railway.com/deploy/anythingllm-rag-workspace) - Self-hosted AI workspace: chat with your docs, agents, multi-user
- [LiteLLM Proxy — Self-Hosted AI Gateway for 100+ LLMs](https://railway.com/deploy/litellm-ai-gateway) - Self-host an AI gateway for 100+ LLMs with spend tracking & budgets.
- [SurfSense](https://railway.com/deploy/surfsense) - Open-source AI research agent (NotebookLM alternative)
- [Bytebot — Self-Hosted AI Desktop Agent with Its Own](https://railway.com/deploy/bytebot-ai-desktop-agent) - Self-host Bytebot: an AI agent with its own full Linux desktop. Any LLM.
- [Hermes Agent + Honcho Memory](https://railway.com/deploy/hermes-agent-honcho-memory) - Deploy and Host Hermes Agent + Honcho Memory with Railway
- [Chat UI Pro Stack: Self Hosted HuggingChat with Bifrost and Mongo](https://railway.com/deploy/chat-ui-pro-stack-self-hosted-huggingcha) - The open chat app behind HuggingChat, on any model, with its own database.
- [MetaMCP](https://railway.com/deploy/metamcp) - Self-hosted MCP gateway — aggregate all your MCP servers in one place.
- [Goose [Updated Sep '26]](https://railway.com/deploy/goose) - Goose [Sep '26] (Extensible AI Agent that Codes, Runs & Tests) Self Host
- [Odysseus (PewDiePie) [Oct'26]](https://railway.com/deploy/odysseus-ai-workspace) - Self-hosted AI workspace with memory, RAG & web search. No ChatGPT fees.
- [Hermes Support Engine](https://railway.com/deploy/hermes-support-engine) - Self-host an AI support agent with persistent memory. No per-seat fees.
- [LightRAG | Open Source Knowledge Graph RAG on Railway](https://railway.com/deploy/light-rag) - Self host LightRAG. Graph-enhanced retrieval with Web UI & multi LLMs
- [robinwiki](https://railway.com/deploy/robinwiki) - Fullstack deployment of the Robin Second brain and Wiki
- [Tarsee](https://railway.com/deploy/tarsee) - 24/7 Claude Code agent — web, Telegram, Discord, voice. No API keys.
- [Ferro Labs AI - SQLite storage](https://railway.com/deploy/ferro-labs-ai-sqlite-storage) - Unified AI Gateway - All your traffic. One gateway - SQLite storage.
- [Ragflow v0.26.4 (Last Version)](https://railway.com/deploy/ragflow-with-infinity-working) - ragflow stack lowest consumption with Infinity, 100% work, RAGFlow rag
- [Agent Gateway](https://railway.com/deploy/agent-gateway) - An open-source agentic proxy for AI agents and MCP servers.
- [Web Tools (Open Source Alternative To Firecrawl, Linkup, Tavily, Exa or Bright Data)](https://railway.com/deploy/web-tools) - Power your AI apps with the world's most accurate open source web tools
- [OpenClaw (Secure)](https://railway.com/deploy/openclaw-secure) - OpenClaw template with persistent secure storage for skills and context
- [nanobot](https://railway.com/deploy/nanobot) - Ultra-Lightweight Personal AI Assistant
- [Open Source Intelligence Command Center](https://railway.com/deploy/open-source-intelligence-command-center) - Plots conflicts, security events, threat indicators on an interactive map
- [Mistral AI](https://railway.com/deploy/mistral-ai) - [Jun '26] Self-host Mistral models locally, using Ollama+OpenWebUI.
- [Open Notebook](https://railway.com/deploy/open-notebook) - Private AI research workspace with chat and podcast generation.
- [I Give Up](https://railway.com/deploy/i-give-up) - AI video generator for Reels/TikTok/Shorts using Google Gemini & Veo
- [AnythingLLM](https://railway.com/deploy/anythingllm-on-railway-or-1-click-llm-ap) - [Jun '26] One click setup for private LLM chat and agent workflows
- [Flowise | Build AI Agents [Postgres Included]](https://railway.com/deploy/flowise-or-build-ai-agents-w-db) - [Jun '26] Low-code developer tool for building LLM Applications
- [FastMCP](https://railway.com/deploy/fastmcp) - Deploy and Host FastMCP with Railway
- [AutoMem - AI Memory Service](https://railway.com/deploy/automem-ai-memory-service) - Graph-vector memory service for AI assistants: https://automem.ai/
- [docling_railway](https://railway.com/deploy/doclingrailway) - Docling + authentication + IP filtering + Gzip + Autosleep configured
- [KittenTTS (Open-Source Text-to-Speech & Voice AI Tool)](https://railway.com/deploy/kittentts) - Kitten TTS [Sep ’26] (Generate Natural AI Voices Instantly) Self Host
- [Langfuse | Open Source LangSmith Alternative for LLM Observability](https://railway.com/deploy/YJ_Ivb) - Self-hosted LLM tracing, prompt management and evaluations
- [Blinko Official ](https://railway.com/deploy/InjVjN) - An open-source, self-hosted personal AI note tool prioritizing privacy
- [letta](https://railway.com/deploy/jgUR1t) - Deploy a Letta server (uses the latest Letta Docker image)
- [LibreChat](https://railway.com/deploy/HxvQtm) - LibreChat is a free, open source AI chat platform.
- [chat2api](https://railway.com/deploy/Q9R9Rg) - A service that can convert ChatGPT on the web to OpenAI API format.
- [Lobe Chat](https://railway.com/deploy/lobe-chat) - one-click deploy your private ChatGPT/Cluade/LLMs chatbot application
- [RSSBox](https://railway.com/deploy/KnVkVX) - Smarter RSS management, more efficient information access
- [Paperclip AI Teams](https://railway.com/deploy/paperclip-ai-teams) - Run AI agent teams (Claude Code, Codex). Owner seeded, invite-only sign-up.
- [OmniRoute Hardened AI Gateway](https://railway.com/deploy/omniroute-hardened-ai-gateway) - OmniRoute AI gateway pinned past CVE-2026-88062, login + API keys enforced
- [Bifrost](https://railway.com/deploy/bifrost-3) - Bifrost AI gateway locked down: dashboard login, keys required, providers
- [SurfSense Self-Hosted Web](https://railway.com/deploy/surfsense-self-hosted-web) - SurfSense NotebookLM alternative with local embeddings and signup gate
- [WeKnora RAG + ParadeDB](https://railway.com/deploy/weknora-rag-paradedb) - Tencent WeKnora RAG and agent platform with ParadeDB, signup closed
- [Scrapling MCP Server + Chromium](https://railway.com/deploy/scrapling-mcp-server-chromium) - Scrapling web scraping MCP server with Chromium and bearer token auth
- [QwenPaw Private Assistant](https://railway.com/deploy/qwenpaw-private-assistant) - QwenPaw personal assistant: login on, shell tool off, one volume
- [MemPalace Remote MCP Hub](https://railway.com/deploy/mempalace-remote-mcp-hub) - Shared AI memory over MCP: token auth, local embeddings, no API keys
- [Open Muse](https://railway.com/deploy/openmuse-template) - Self-hosted OpenMuse: personal AI agent with browser, terminal, and files
- [Ollama](https://railway.com/deploy/ollama-1) - Ollama with an API key in front, models on a volume, pinned release
- [SearXNG MCP](https://railway.com/deploy/searxng-mcp) - Web search for AI agents over MCP, with a private SearXNG and no API key
- [Open WebUI](https://railway.com/deploy/open-webui-8) - Open WebUI on Postgres with pgvector, admin account set at deploy time
- [LibreChat](https://railway.com/deploy/librechat-2) - LibreChat on a pinned release: admin set at deploy, file search, Mongo
- [LobeHub](https://railway.com/deploy/lobehub-1) - LobeHub, formerly LobeChat: AI chat and agents with Postgres and storage
- [Hermes Agent with Hindsight Memory](https://railway.com/deploy/hermes-agent-with-hindsight-memory) - Hermes Agent with self-hosted Hindsight memory. One OpenRouter key for both
- [OpenCode](https://railway.com/deploy/opencode-web) - OpenCode's web UI with a password, your projects and logins on a volume
- [Onyx Lite](https://railway.com/deploy/onyx-lite) - Onyx Lite - Open Source AI Platform - AI Chat with advanced features
- [FreeLLMAPI [Updated Oct'26]](https://railway.com/deploy/free-llm-api-gateway) - Self-host FreeLLMAPI — OpenAI-compatible router, failover, encrypted keys
- [Playwright MCP](https://railway.com/deploy/playwright-mcp) - Playwright MCP headless browser for AI agents, behind a token gateway.
- [sundhedmcp](https://railway.com/deploy/sundhedmcp) - Let Claude read your own sundhed.dk record after a MitID login. Read-only.
- [openmaic](https://railway.com/deploy/openmaic-1) - Multi-agent AI classroom — one-click lessons from any topic. Self-hosted.
- [openclaw-v2](https://railway.com/deploy/openclaw-v2) - Self-host OpenClaw 2.0 Gateway and Control UI
- [Rakazo Suite](https://railway.com/deploy/rakazo-suite) - Rakazo persistent cloud agent (Grok Bot Alternative). Non forked.
- [Canvas Viewer MCP Server](https://railway.com/deploy/canvas-viewer-mcp-server) - Give your AI agent read-only access to your Canvas account
- [new-api-template](https://railway.com/deploy/new-api-template) - new-api LLM gateway on Railway: MySQL+Redis, auto secrets, persistence
- [ContextForge](https://railway.com/deploy/contextforge) - Gateway and registry that unifies MCP servers behind one endpoint
- [Open WebUI + Ollama](https://railway.com/deploy/openwebui-ollama-private-ai) - Self-host a private, login-gated chat UI — your data stays yours [Oct'26]
- [DeskcommCRM](https://railway.com/deploy/deskcommcrm) - WhatsApp sales CRM with AI agents and a self-hosted Supabase, all in one
- [Engram](https://railway.com/deploy/engram-cloud) - Persistent, searchable memory that AI coding agents read and write
- [LightRAG Server + Web UI](https://railway.com/deploy/lightrag-server-web-ui) - Graph-based RAG server with web UI, auth on, file storage on a volume
- [Kokoro TTS API (CPU)](https://railway.com/deploy/kokoro-tts-api-cpu) - OpenAI-compatible Kokoro text-to-speech API on CPU, model baked in
- [LiteLLM Gateway + Postgres](https://railway.com/deploy/litellm-gateway-postgres) - OpenAI-compatible LLM gateway with Postgres, virtual keys, spend tracking
- [Workmates](https://railway.com/deploy/workmates) - Deploy and host Workmates AI agents with desktop and mobile access.
- [Soulacy Personal](https://railway.com/deploy/soulacy-personal) - Private AI agent gateway with generated secrets and secure iPhone pairing.
- [phoenix (by arize): LLM Evals and Observability - Open Source](https://railway.com/deploy/phoenix-by-arize-llm-evals-and-observabi) - Observe and Evaluate your models and agents. Self-Host it.
- [hermes agent: easy multi-model option for codex, open claw, claude code, codex and more](https://railway.com/deploy/hermes-agent-easy-multi-model-option-for) - The simplest curated setup for an Hermes Agent. Curated by impacte.tech
- [RAGFlow CPU](https://railway.com/deploy/ragflow-cpu) - Document ingestion and retrieval with private databases and CPU workers.
- [LobeHub (Lobe Chat) | Team AI Chat with Postgres & File Storage](https://railway.com/deploy/lobehub-chat) - LobeHub server edition: team AI chat, files, knowledge base. Zero config.
- [ollama + openweb-ui: run your own open models, own your (free) tokens](https://railway.com/deploy/ollama-openweb-ui-run-your-own-open-mode) - Ollama + OpenWeb-UI + Real, Working, Local Models
- [property-radar](https://railway.com/deploy/property-radar) - A self-hosted MCP server for discovering properties for sale or rent.
- [Agno](https://railway.com/deploy/agno) - A production-ready agent runtime with API, MCP, memory and observability.
- [Cognee — Self-Hosted AI Memory with MCP](https://railway.com/deploy/cognee-ai-memory) - Self-host Cognee — persistent agent memory over MCP, knowledge graph
- [TencentDB Agent Memory](https://railway.com/deploy/tencentdb-agent-memory) - TencentDB Agent Memory is a team-level memory hub for AI Agents.
- [Bolt DIY | (Just Updated) bolt.new Alternative That Actually Boots](https://railway.com/deploy/bolt-diy-or-just-updated-boltnew-alterna) - bolt.new alternative that boots, with your API keys behind a password
- [APITable](https://railway.com/deploy/apitable) - API-oriented low-code platform | Airtable alternative
- [Sim Studio](https://railway.com/deploy/sim-studio-1) - Visual builder for AI agents and automated workflows
- [Spacebot](https://railway.com/deploy/spacebot-3) - Deploy Spacebot with Caddy HTTP Basic Auth and persistent storage.
- [AnythingLLM](https://railway.com/deploy/anythingllm-chat) - Chat with your own documents using the AI model you choose
- [Strix AI Pentest Agent](https://railway.com/deploy/strix-ai-pentest-agent) - Autonomous AI penetration-testing agent as a headless Railway worker.
- [New API | (Just Updated) LLM Gateway Nobody Else Can Claim](https://railway.com/deploy/new-api-or-just-updated-llm-gateway-nobo) - Multi-provider LLM gateway; admin seeded at boot, sessions survive redeploy
- [Dify | (Just Updated) LLM App Builder Whose Plugin System Actually Works](https://railway.com/deploy/dify-or-just-updated-llm-app-builder-who) - Dify 1.16.1, 6 services not 11, plugin marketplace verified working
- [Multica | Open Source Linear for AI Coding Agents](https://railway.com/deploy/multica-coding-agents) - Issue tracker where you assign tasks to Claude Code, Codex, Cursor
- [Bifrost](https://railway.com/deploy/bifrost-gateway) - Self-host Bifrost, the Go AI gateway for 20+ model providers
- [Whisper STT API | (Just Updated) Transcription That Answers the First Request](https://railway.com/deploy/whisper-stt-api-or-just-updated-transcri) - OpenAI-compatible speech to text, model baked in, threads sized to the box
- [Presenton | Open Source AI Presentation Generator](https://railway.com/deploy/presenton) - Self-hosted AI presentation & slide generator with persistent storage
- [Executor | Open Source MCP Gateway](https://railway.com/deploy/executor-selfhost) - Self-hosted MCP gateway: connect any AI agent to your APIs and tools
- [LangWatch agent testing](https://railway.com/deploy/langwatch-agent-testing) - Private LangWatch agent testing with durable traces and evaluations.
- [ContextForge MCP gateway](https://railway.com/deploy/contextforge-mcp-gateway) - Secure MCP gateway and registry with PostgreSQL, Redis, and admin UI.
- [Promptfoo evaluation](https://railway.com/deploy/promptfoo-evaluation) - Private Promptfoo evaluation and red teaming with persistent project data.
- [BentoML API starter](https://railway.com/deploy/bentoml-api-starter) - Production-minded BentoML API starter with typed batch inference.
- [Aegra agent backend](https://railway.com/deploy/aegra-agent-backend) - Durable agent backend with LangGraph APIs, pgvector, and Redis workers.
- [OpenWork](https://railway.com/deploy/openwork) - OpenWork w/ OpenCode, persistent storage, remote access. openwork open work
- [Mem0 self-hosted](https://railway.com/deploy/mem0-self-hosted-1) - Private AI memory platform with dashboard and persistent pgvector storage.
- [postiz Pro Stack One-click Deploy With S3](https://railway.com/deploy/postiz-pro-stack-one-click-deploy) - postiz, Temporal, Redis & S3 social media automation. Postiz
- [Firecrawl](https://railway.com/deploy/firecrawl-scraper-api) - Host Firecrawl [Oct'26] — scrape & crawl any site to clean Markdown
- [9Router | Self-Hosted AI Gateway, Nothing to Configure](https://railway.com/deploy/9router-or-self-hosted-ai-gateway-nothin) - Self-hosted AI gateway. 40+ providers, one endpoint, nothing to configure.
- [LibreTranslate [Updated Oct'26]](https://railway.com/deploy/libretranslate-translation-api) - Self-host translation — REST API, 30+ languages, no per-word fees
- [Hermes Agent | OpenClaw Alternative, Nothing to Configure](https://railway.com/deploy/hermes-agent-or-openclaw-alternative-not) - Self-improving Nous Research agent. Nothing to fill in, gated dashboard.
- [Airi](https://railway.com/deploy/airi) - AIRI Web, a self-hosted AI companion with Live2D, VRM and voice. airi
- [Open WebUI](https://railway.com/deploy/open-webui-4) - Open WebUI — ChatGPT-style UI for Ollama and OpenAI-compatible models
- [ion-agent](https://railway.com/deploy/ion-agent) - Deploy and Host ion-agent with Railway
- [Coder — Self-Hosted Dev Environments Control Plane](https://railway.com/deploy/coder-control-plane-postgres) - Open-source Codespaces alternative — control plane + Postgres
- [systemprompt.io: the self-owned AI control plane](https://railway.com/deploy/systempromptio-the-self-owned-ai-control) - Stop renting your AI. The Rust control plane you own.
- [Convex — Self-Hosted Reactive Backend & Firebase Alternative](https://railway.com/deploy/convex-reactive-backend) - Self-host Convex: reactive backend with live queries. No per-dev cloud fee.
- [OpenClaw [Updated Sep '26]](https://railway.com/deploy/openclaw-updated-jul-26) - OpenClaw [Sep '26] (Self-Hosted AI Assistant & Multi-Channel Bot) Self Host
- [hajimi - Gemini API Proxy](https://railway.com/deploy/hajimi-gemini-api-proxy) - OpenAI-compatible Gemini API proxy with key rotation & Vertex AI
- [LiteLLM & One API Alternative — CoAI Gateway](https://railway.com/deploy/litellm-and-one-api-alternative-coai-gat) - Self-hosted AI & LLM gateway with UI,analytics,billing and API Managment.
- [Vane — Self-Hosted AI Search, Formerly Perplexica](https://railway.com/deploy/vane-ai-search) - Self-host Vane (ex-Perplexica): AI web search with cited answers.
- [Latitude](https://railway.com/deploy/latitude) - Open-source AI Agent Monitoring Platform
- [Odysseus](https://railway.com/deploy/odysseus) - odysseus from pewdiepie. Self-hosted AI workspace. odyseus
- [Self-Host OpenAI-Compatible STT & TTS on Railway — Speaches](https://railway.com/deploy/speaches-stt-tts-api) - Self-host OpenAI-compatible STT & TTS. No per-minute fee. Audio stays local
- [Flowise](https://railway.com/deploy/flowise-ai-agent) - [Jul'26] Lightweight Flowise with web UI, auth, and persistent storage 🤖
- [R2R | The most advanced AI retrieval system by SciPhi](https://railway.com/deploy/r2r) - Self-Host R2R. Agentic RAG, hybrid search, document management with API
- [Flowise – Self-Hosted AI Agent Builder](https://railway.com/deploy/flowise-self-hosted-ai-agent-builder) - Self-hosted Flowise: Build AI agents, chatbots & RAG pipelines visually.
- [Khoj | Open Source Personal AI Assistant](https://railway.com/deploy/khoj) - Self Host Khoj. AI second brain with document search and custom agents
- [LocalAI | Private OpenAI-Compatible API](https://railway.com/deploy/localai-openai-compatible-api) - Self-host LocalAI on Railway, run LLMs, images, and audio locally
- [worldmonitor](https://railway.com/deploy/worldmonitor) - Real-time global intelligence dashboard with interactive map
- [archestra](https://railway.com/deploy/archestra) - MCP-native AI platform with web UI and secure agent orchestration
- [goclaw](https://railway.com/deploy/goclaw) - Multi-tenant Go AI agent platform with dashboard, memory, and 20+ providers
- [Promptfoo](https://railway.com/deploy/promptfoo) - An open-source tool for testing, evaluating, and red-teaming LLM apps.
- [Markitdown - MCP + REST](https://railway.com/deploy/markitdown-mcp-rest) - Deploy and Host markitdown-mcp-rest with Railway
- [NanoClaw (OpenClaw Lightweight Alternative)](https://railway.com/deploy/nanoclaw) - OpenClaw alternative with multi-channel, memory & Agents SDK built in
- [PicoClaw | Lightweight Alternative of OpenClaw | Self Host](https://railway.com/deploy/picoclaw-or-lightweight-alternative-of-o) - [May '26] One-click deploy PicoClaw. Any LLM. Any messaging platform.
- [ZeroClaw (CLI-only)](https://railway.com/deploy/zeroclaw-cli-only) - Manage your zeroclaw directly in terminal with railway-cli ssh
- [Coworker](https://railway.com/deploy/coworker) - Array Ventures: Coworker Agent
- [ChromaDB](https://railway.com/deploy/chromadb-1) - Official Docker image - AI vector database for RAG, LLMs & hybrid search
- [MCP Inspector](https://railway.com/deploy/mcp-inspector) - Inspect and debug MCP (Model Context Protocol) servers
- [Perplexica](https://railway.com/deploy/perplexica) - Open source alternative to Perplexity AI
- [MaxKB](https://railway.com/deploy/maxkb) - Open-source platform for building enterprise-grade agents
- [Sim AI](https://railway.com/deploy/sim-ai) - Deploy and Host Sim AI with Railway
- [Fire Enrich](https://railway.com/deploy/fire-enrich) - Fire Enrich - AI-Powered Data Enrichment Tool
- [Letta [Updated Sep ’26] (Open-Source AI Agent Framework & Orchestration Platform)](https://railway.com/deploy/letta) - Letta [Sep ’26] (Build, Deploy & Manage AI Agents Seamlessly) Self Host
- [Qwen3-0.6b](https://railway.com/deploy/qwen3-06b) - Deploy and Host Qwen3-0.6b on Railway.
- [Open Loveable (MendableAI)](https://railway.com/deploy/open-loveable-mendableai) - Deploy and Host open-loveable (MendableAI) with Railway
- [Weaviate](https://railway.com/deploy/weaviate) - Weaviate DB with secure, password-protected default configuration.
- [Node-RED Context AI](https://railway.com/deploy/node-red-mcp) - AI workflows in Node-RED using Model Context Protocol (MCP)
- [Evo AI](https://railway.com/deploy/mpbtSK) - Open-source platform to create and manage AI agents
- [NSFW Detection API](https://railway.com/deploy/Tlmof_) - FastAPI NSFW Detection API powered by Hugging Face.
- [Langfuse v3.225.11 (Production ready AI Observability)](https://railway.com/deploy/kUsf_w) - Open-source LLM platform for prompt management, tracing, evals & metrics
- [big-AGI](https://railway.com/deploy/BOIjs9) - powered up ChatGPT - Gennerative AI Suite with advanced multi-modal tools
- [firecrawl](https://railway.com/deploy/AIaBEM) - firecrawl api server + worker without auth, works with dify
- [Suno-API](https://railway.com/deploy/IdlBP8) - REST API for Suno AI for interacting with Suno AI's music generator.
- [SoraFlows](https://railway.com/deploy/39bNf0) - SoraFlows is a browser interface for OpenAI Sora, generate text-to-video. 
- [Argilla](https://railway.com/deploy/KNxfha) - The open-source feedback platform for LLMs
- [Chroma](https://railway.com/deploy/chroma-1) - Chroma vector database for AI apps - embeddings, RAG & semantic search
- [Whistle](https://railway.com/deploy/whistle-stt-api) - Tiny on-device speech-to-text API. OpenAI-compatible
- [Jev Lite](https://railway.com/deploy/jev-lite-1) - Self-hosted web console + API proxy for TypeSafe Jev decision model
- [opendots](https://railway.com/deploy/opendots-2) - Persistent AI coworker workspace with chat, research, pages, and calls
- [AI Memory](https://railway.com/deploy/ai-memory) - One-click deploy: shared AI agent memory with MCP and authentication.
- [nanollm](https://railway.com/deploy/nanollm) - Lightweight LLM gateway with web admin and private persistent SQLite.
- [Tcpcore demo](https://railway.com/deploy/tcpcore-demo) - Runtime governance for every agent API call
- [MCPJam Inspector](https://railway.com/deploy/mcpjam-inspector) - Self-host MCPJam Inspector: test and debug MCP servers in a browser
- [OpenMuse](https://railway.com/deploy/openmuse-1) - CopilotKit's personal agent with a browser, files and tasks, on Postgres
- [AionUi | Cowork AI Agent Workspace in Your Browser](https://railway.com/deploy/aionui) - AionUi WebUI: chat, files, scheduled tasks, teams of AI agents, any model.
- [Hermes Agent | Official Image, Web Dashboard, Browser Chat, Telegram](https://railway.com/deploy/hermes-agent-dashboard) - Nous Research's self-improving agent: web dashboard, browser chat, Telegram
- [OpenClaw (Self-Hosted)](https://railway.com/deploy/openclaw-self-hosted) - Self-host OpenClaw with Chromium, persistent memory and the Control UI.
- [Open WebUI | Self-Hosted LLM Chat with a Real Volume](https://railway.com/deploy/open-webui-or-self-1) - Self-hosted chat UI for LLMs, on a pinned release with a persistent volume
- [bolt.diy | Browser AI App Builder, Pinned and in Production Mode](https://railway.com/deploy/boltdiy-or-browser-1) - bolt.diy on Railway: pinned build, production mode, no keys demanded
- [Nanoclaw](https://railway.com/deploy/nanoclaw-1) - Personal Claude AI assistant for Slack, Telegram, Discord, WhatsApp, Gmail
- [Whisper ASR](https://railway.com/deploy/whisper-asr) - Whisper ASR 1.10: speech-to-text API with faster-whisper, private network.
- [SillyTavern [Updated Sep '26]](https://railway.com/deploy/sillytavern-llm) - SillyTavern — LLM Frontend, Auto-Secured Admin Password
- [Crawl4AI](https://railway.com/deploy/crawl4ai-1) - Open-source LLM-friendly web crawler with REST API, playground and MCP.
- [Hermes Agent](https://railway.com/deploy/hermes-agent-official-image) - Nous Research's Hermes Agent: memory on a volume, API with a key, Telegram
- [OpenHands + DeepSeek AI Coding Agent](https://railway.com/deploy/openhands-deepseek-ai-coding-agent) - AI coding agent with DeepSeek, GitHub integration, and persistent workspace
- [Sim](https://railway.com/deploy/sim) - Sim 0.8.59: open-source visual builder for AI agent workflows.
- [MLflow](https://railway.com/deploy/mlflow-1) - MLflow 3.16 tracking server with login, Postgres and artifact storage.
- [llama.cpp Server](https://railway.com/deploy/llamacpp-server) - llama.cpp server with an OpenAI-compatible API running GGUF models on CPU.
- [OpenMuse](https://railway.com/deploy/openmuse) - OpenMuse personal AI assistant with browser, tasks, goals and Postgres
- [Supermemory with Cloudflare Zero Trust](https://railway.com/deploy/supermemory-with-cloudflare-zero-trust) - Supermemory memory engine with Cloudflare Zero Trust Access protection.
- [Onyx Lite Chat + Postgres](https://railway.com/deploy/onyx-lite-chat-postgres) - Onyx Lite self hosted AI chat with Postgres, four services
- [RAGFlow Document RAG Stack](https://railway.com/deploy/ragflow-document-rag-stack) - RAGFlow document RAG with Elasticsearch, MySQL, MinIO and Valkey
- [Text Embeddings Inference + Model Cache](https://railway.com/deploy/text-embeddings-inference-model-cache) - Hugging Face TEI embeddings API with API key auth and model cache
- [Speaches Audio API + Model Cache](https://railway.com/deploy/speaches-audio-api-model-cache) - OpenAI-compatible speech to text and text to speech with a model cache
- [LibreChat + Meilisearch + RAG](https://railway.com/deploy/librechat-meilisearch-rag) - LibreChat with Meilisearch search and a keyless RAG API for file chat
- [Bifrost Gateway (Go)](https://railway.com/deploy/bifrost-gateway-go) - Go LLM gateway with dashboard auth on, virtual keys, budgets and fallbacks
- [Langflow](https://railway.com/deploy/kit-langflow-build-0919) - Langflow: visual builder for AI agents and RAG flows. Requires Hobby plan
- [Omnigent](https://railway.com/deploy/omnigent) - Self-hosted AI agent meta-harness (Claude Code, Codex); admin seeded.
- [Qdrant Lite](https://railway.com/deploy/qdrant-lite) - Deploy Qdrant vector database with persistent storage and web dashboard
- [Unorouter](https://railway.com/deploy/unorouter) - Self-hosted AI gateway + storefront. One key for 100+ AI providers.
- [PaddleOCR](https://railway.com/deploy/paddleocr) - OCR API for images and PDFs, 100+ languages, CPU-only
- [AgentMarginRouter](https://railway.com/deploy/agentmarginrouter-1) - Agent Margin Router – x402 Data API for AI Agents and Automations
- [QwenPaw](https://railway.com/deploy/qwenpaw-1) - Self-hosted AI assistant with memory, skills and browser automation
- [Octop](https://railway.com/deploy/octop) - Self-hosted multi-user AI assistant with agents, knowledge base and cron
- [Ubuntu OpenClaw Desktop (GUI)](https://railway.com/deploy/openclaw-desktop-gui) - OpenClaw on an Ubuntu desktop: watch its Chrome, sign in to sites for it
- [HolyClaude Workstation](https://railway.com/deploy/holyclaude-workstation) - AI coding workstation: Claude Code, web UI, headless browser, 8 AI CLIs
- [ContextForge MCP Gateway](https://railway.com/deploy/contextforge-mcp-g-1) - Self-hosted IBM registry, proxy and admin UI for MCP servers and REST APIs
- [codeg](https://railway.com/deploy/codeg) - Multi-agent AI coding workspace for Claude Code, Codex and others
- [Rakazo (Single Container)](https://railway.com/deploy/rakazo-single-container) - Persistent AI teammates with memory and routines. BYO model key
- [GoModel](https://railway.com/deploy/gomodel) - AI gateway with PostgreSQL, Redis caching, and OpenAI/Anthropic APIs.
- [GoModel Lite](https://railway.com/deploy/gomodel-lite) - Single-instance AI gateway with SQLite and OpenAI/Anthropic APIs.
- [WeKnora](https://railway.com/deploy/weknora) - Tencent's open-source RAG knowledge base and agent platform.
- [Pipecat](https://railway.com/deploy/pipecat) - Browser voice AI with OpenAI, Gemini, Grok, OpenRouter, or Ollama.
- [Litellm-Novita-ai](https://railway.com/deploy/litellm-novita-ai) - Litellm setup connected to novita.ai to execute open source models
- [Acontext Agent Context Platform](https://railway.com/deploy/acontext-agent-context-platform) - Agent context management with sessions, queues and object storage.
- [Marker PDF to Markdown API](https://railway.com/deploy/marker-pdf-to-markdown-api) - Authenticated CPU API for converting digital PDFs to Markdown.
- [OpenRAG CPU Stack](https://railway.com/deploy/openrag-cpu-stack) - RAG workspace with OpenSearch, Langflow and CPU document processing.
- [BettaFish Research Reports](https://railway.com/deploy/bettafish-research-reports) - Multi-agent research and reports with PostgreSQL and persistent files.
- [Coze Studio](https://railway.com/deploy/coze-studio) - Agent and workflow studio with SQL, vector search and object storage.
- [AgentScope Service API](https://railway.com/deploy/agentscope-service-api) - AgentScope API with Redis state and persistent agent workspaces.
- [Xberg / Kreuzberg Document API](https://railway.com/deploy/xberg-kreuzberg-do-1) - CPU document extraction API with authenticated HTTPS access.
- [FastGPT Core + RAG](https://railway.com/deploy/fastgpt-core-rag) - FastGPT chat and RAG with MongoDB, pgvector, Redis and object storage.
- [JupyterLab](https://railway.com/deploy/jupyterlab-1) - Persistent JupyterLab with scientific Python and generated token login.
- [Gitingest](https://railway.com/deploy/gitingest-llm) - Turns a Git repository into one text file you can paste into an LLM
- [Leon](https://railway.com/deploy/leon) - Alexa Alternative: Personal assistant that runs on your own server
- [PicoClaw](https://railway.com/deploy/picoclaw-agent) - Tiny self-hosted AI assistant with file, shell and cron tools
- [a2a-starter-agent](https://railway.com/deploy/a2a-starter-agent) - Minimal, LLM-agnostic A2A protocol agent for any A2A client
- [MindsDB](https://railway.com/deploy/mindsdb-sql) - Query engine that runs SQL and AI across all your data sources
- [Pydantic AI](https://railway.com/deploy/pydantic-ai) - A typed AI agent runtime with web chat and persistent memory.
- [AI Proxy](https://railway.com/deploy/ai-proxy) - A high-performance AI gateway for OpenAI, Claude, Gemini & more.
- [Aegra](https://railway.com/deploy/aegra) - A self-hosted backend for building and running LangGraph AI agents.
- [Opik](https://railway.com/deploy/opik) - Open-source LLM observability: evaluation, tracing and guardrails
- [DeepTutor - personalized AI tutoring](https://railway.com/deploy/deeptutor-personalized-ai-tutoring) - Deploy DeepTutor for personalized, source-grounded AI tutoring.
- [Mem0](https://railway.com/deploy/mem0-ai) - Long-term memory for AI agents that remembers across conversations
- [RAGFlow](https://railway.com/deploy/ragflow-ai) - Turns your documents into a searchable, citable knowledge base
- [Onyx](https://railway.com/deploy/onyx-ai) - AI chat and search over your company's documents and tools
- [ZeroClaw AI Assistant](https://railway.com/deploy/zeroclaw-ai-assistant) - Personal AI assistant with dashboard, persistent storage, and healthchecks.
- [OpenClawfice](https://railway.com/deploy/openclawfice) - Turn your OpenClaw agents into pixel-art NPCs and watch them work live.
- [Claw3D](https://railway.com/deploy/claw3d) - A 3D virtual office for OpenClaw, Hermes, and custom AI agent runtimes.
- [ClawRouter](https://railway.com/deploy/clawrouter) - Self-hosted OpenAI-compatible LLM router with pay-per-call crypto billing
- [Skyvern](https://railway.com/deploy/skyvern) - AI agents that drive a real browser to fill forms and scrape sites.
- [AutoGen Studio](https://railway.com/deploy/autogenstudio) - Visual tool for building and testing teams of AI agents
- [Multica](https://railway.com/deploy/multica-railway-template) - Multica on Railway with automatic URL/daemon setup, OpenCode, and Postgres.
- [Weaviate](https://railway.com/deploy/weaviate-db-1) - Open-source vector database for semantic search, RAG, and AI apps.
- [Qdrant](https://railway.com/deploy/qdrant-db) - Vector database for AI search, embeddings, RAG, and semantic retrieval.
- [wemux-community](https://railway.com/deploy/wemux-community) - Self-hostable AI agent collaboration platform
- [FastMCP](https://railway.com/deploy/fastmcp-server) - A server that gives AI assistants tools and memory
- [DeerFlow](https://railway.com/deploy/deer-flow) - AI assistant that researches topics and writes reports for you
- [Letta](https://railway.com/deploy/letta-agents) - AI agents that remember you and get better over time
- [LightRAG](https://railway.com/deploy/lightrag-rag) - Turns your documents into a searchable knowledge graph you can query
- [Evo AI](https://railway.com/deploy/evo-ai) - Self-hosted AI agent platform with LangGraph workflows and MCP tools
- [LibreTranslate](https://railway.com/deploy/libretranslate-server) - Translates text and documents between about 50 languages
- [Ultimate Voice Cloner | (Just Updated) 4GB CPU Voice Cloning, MOSS-TTS, Answers Cold](https://railway.com/deploy/ultimate-voice-cloner-or-just-updated-ze) - Zero-shot voice cloning on CPU, MOSS-TTS-Nano. One POST route, key-gated
- [AutoMem](https://railway.com/deploy/automem) - Long-term memory for AI assistants that recalls past decisions
- [personal-llm-gateway](https://railway.com/deploy/personal-llm-gateway) - Personal LiteLLM gateway with CLIProxyAPI accounts and virtual keys.
- [OpenViking](https://railway.com/deploy/openviking-aug-26-agent-memory) - Persistent memory and RAG for AI agents. Store context across sessions.
- [Flowise](https://railway.com/deploy/flowise-railway) - AI agent builder. Visual LLM app & AI agent builder [with Worker]
- [Flowise — AI Agent Builder](https://railway.com/deploy/flowise-visual-ai-builder) - Build AI agents visually with a powerful low-code workflow platform.
- [Hermes Agent on NixOS | (Just Updated) Self-Improving Agent, Memory Persists](https://railway.com/deploy/hermes-agent-on-nixos-or-just-updated-se) - Self-improving agent on nix. Memory and skills survive redeploys. 1 GB min.
- [Weaviate](https://railway.com/deploy/weaviate-db) - Pinecone Alternative. Open-source vector database for semantic search & RAG
- [Magic Reactions for ChatGPT](https://railway.com/deploy/magic-reactions-for-chatgpt) - Private GIF and sticker reactions for ChatGPT, with OAuth and storage.
- [NOFX | (Just Updated) AI Trading Platform That Actually Boots](https://railway.com/deploy/nofx-or-just-updated-ai-trading-platform) - AI trading platform that boots, is pre-claimed, and keeps its keys
- [kandev](https://railway.com/deploy/kandev) - Coordinate AI coding workflows from a secure, persistent web UI
- [Contxtyfy Command Centre](https://railway.com/deploy/contxtyfy-command-centre) - Contxtyfy is a personal AI Command Centre and Context Knowledge Graph
- [Buzz | Open Source Slack Alternative with AI Agents](https://railway.com/deploy/buzz-relay-workspace) - Self-hosted team chat with AI agents, git repos and QR pairing
- [LibreChat | Open Source ChatGPT Alternative](https://railway.com/deploy/librechat-ai-chat) - 5-service LibreChat with search and RAG. ~1.5 GB RAM, about $25/month
- [LiteLLM | Open Source OpenAI-Compatible LLM Gateway](https://railway.com/deploy/litellm-llm-gateway) - OpenAI-compatible proxy for 100+ LLM providers, with keys and budgets
- [Agent Control Dev](https://railway.com/deploy/agent-control-dev) - Deploy Agent Control server - guardrails, API, and UI
- [Mastra Durable Agents](https://railway.com/deploy/mastra-durable-agents) - Durable AI agents with resumable streams via Redis and Inngest.
- [mlflow: Agent Tracking & Tracing | Real Observability for your AI Agents](https://railway.com/deploy/mlflow-agent-o11y-template) - Self-hosted MLflow tracing backend with an instrumented demo agent
- [Odysseus (w/ Search + RAG)](https://railway.com/deploy/odysseus-w-search-rag) - Self-hosted AI workspace: chat, agents, deep research, notes, docs, email
- [Open WebUI — Self-Hosted RAG & Document AI Chat](https://railway.com/deploy/open-webui-rag-documents) - Self-host Open WebUI — chat with your PDFs & docs, privately
- [FlowiseAI [Updated Sep '26]](https://railway.com/deploy/flowiseai) - Flowise [Sep '26] (Visual Builder for LLM Apps & AI Agents) Self Host
- [Docling Serve](https://railway.com/deploy/docling-serve) - API-key-protected CPU document conversion
- [Open WebUI](https://railway.com/deploy/open-webui-ai-chat) - Private AI chat UI for OpenAI, Ollama and any compatible API
- [R2R RAG backend](https://railway.com/deploy/r2r-rag-backend) - Authenticated R2R retrieval with pgvector ingestion and cited RAG.
- [Open Wearables — Unified Wearable Health API](https://railway.com/deploy/open-wearables-health-api) - Self-host one API for Garmin, Whoop, Oura & Fitbit health data
- [Qdrant [Updated Sep '26]](https://railway.com/deploy/qdrant-1) - Qdrant [Sep '26] (Open-Source Vector Database for AI & Search) Self Host
- [Piney - AI Character Studio](https://railway.com/deploy/piney) - SillyTavern Character Card Workstation.
- [Amadeus System](https://railway.com/deploy/amadeussystem) - A Multimodal AI Application Inspired By Steins;Gate 0
- [LibreChat | ChatGPT Alternative, Files + Search That Persist](https://railway.com/deploy/librechat-or-chatgpt-alternative-files-s) - Self-hosted ChatGPT alternative: file RAG with no API key, uploads persist
- [LibreChat](https://railway.com/deploy/librechat-1) - Multi-provider AI chat with agents, files, search, and RAG.
- [LibreTranslate](https://railway.com/deploy/libretranslate-3) - Private machine translation API and web interface with local models.
- [Manifest — LLM Provider Gateway for Agents](https://railway.com/deploy/manifest-llm-router) - Self-host Manifest — one endpoint for keys, subs, local & custom
- [LocalAI — Private OpenAI API for LLM, Image & Audio](https://railway.com/deploy/localai-multimodal-openai-api) - Self-host one OpenAI-compatible API — LLM, images & audio
- [Flowise + n8n — AI Agent Automation Stack](https://railway.com/deploy/flowise-n8n-ai-automation-stack) - Self-host Flowise + n8n — AI agents & workflow automation stack
- [LobeChat — Multi-Model AI Chat with RAG [Postgres]](https://railway.com/deploy/lobechat-server-database-rag) - Self-host LobeChat with history, RAG & 40+ providers — server mode
- [Owyhee-governed OpenClaw agent](https://railway.com/deploy/owyhee-governed-openclaw-agent) - OpenClaw agent with Owyhee governance: custody, approvals, audit trail.
- [MOSS TTS - Voice Cloning API (CPU, OpenAI-compatible)](https://railway.com/deploy/moss-tts-voice-cloning-api-cpu-openai-co) - Voice-cloning TTS with an OpenAI-compatible API. CPU-only, no GPU bill.
- [Codex Pooler](https://railway.com/deploy/codex-pooler) - Self-hosted Codex gateway with pooled accounts and OpenAI-compatible APIs.
- [SenseVoice ASR Server](https://railway.com/deploy/sensevoice-asr-server) - A lightweight, CPU-based ASR service using the SenseVoice model.
- [Lemonade Server](https://railway.com/deploy/lemonade-server-1) - Lemonade — OpenAI-compatible local AI model inference server
- [open-connector](https://railway.com/deploy/open-connector) - connecting 1000+ SaaS providers to AI agents through MCP, HTTP, and OpenAPI
- [OpenAI, Gemini & Claude REST API Gateway](https://railway.com/deploy/gemini-api-gateway-one-api-for-gemini-cl) - Unified REST API Gateway for OpenAI, Gemini, Antigravity and Claude models.
- [Bifrost AI Gateway Stack: LLM Proxy with Caching, Budgets, and Logs](https://railway.com/deploy/bifrost-ai-gateway-stack-llm-proxy-with-) - A fast OpenAI compatible gateway with response caching across every model.
- [Open WebUI [Jul'26]](https://railway.com/deploy/open-webui-6) - Self-hosted AI chat UI for Ollama and OpenAI
- [Mastra Studio Standalone](https://railway.com/deploy/mastra-studio-standalone) - Standalone Mastra Studio UI - connect to any running Mastra server
- [LobeChat + Bifrost | Multi-Provider AI Chat with Gateway](https://railway.com/deploy/lobechat-or-multi-provider-ai-chat-with-) - Deploy and Host LobeChat | Multi-Provider AI Chat with a Built-In Gateway
- [Open WebUI + Bifrost | Self-Hosted ChatGPT for Every LLM](https://railway.com/deploy/open-webui-bifrost-or-self-hosted-chatgp) - Deploy and Host Open WebUI + Bifrost
- [Self-Host Plandex — AI Coding Agent for Large Codebases on Railway](https://railway.com/deploy/plandex-ai-coding-server) - Self-host Plandex: 2M token context, plan version control, BYO API keys.
- [Deploy and Host SnapOtter Image Toolkit on Railway](https://railway.com/deploy/snapotter) - Self-host 52 image tools + local AI. No per-image fees. Images never leave.
- [Open WebUI with Ollama](https://railway.com/deploy/open-webui-with-ollama) - [Jul'26] Open WebUI with Ollama bundled. Download models, chat privately.
- [LiveKit Voice Agent — Self-Hosted WebRTC AI on Railway](https://railway.com/deploy/livekit-voice-agent) - Self-host LiveKit: OpenAI voice agent with WebRTC. No per-minute fees.
- [beevibe](https://railway.com/deploy/beevibe) - Self-hosted agent runtime — pinned to release branch (latest stable tag)
- [Flowise with postgres](https://railway.com/deploy/flowise-with-postgres) - Flowise with Postgres, web UI, auth & persistent file storage. 🚀
- [Flowise + Postgres](https://railway.com/deploy/flowise-postgres) - Flowise AI agents with Postgres, persistent storage, and auth.
- [Ministers Template](https://railway.com/deploy/ministers-template) - Ministers T is a ready-to-run AI Chief of Staff
- [atomic](https://railway.com/deploy/atomic) - Self-hosted, semantically-connected personal knowledge base
- [Plexus](https://railway.com/deploy/plexus) - Unified API gateway for multiple AI providers.
- [manifes](https://railway.com/deploy/manifes) - Smart model router with dashboard, routing tiers, and cost analytics
- [Bytebot - AI desktop agent](https://railway.com/deploy/bytebot-1) - Deploy Bytebot: The AI desktop agent that automates tasks.
- [Cognee AI Memory Platform with MCP](https://railway.com/deploy/cognee-ai-memory-p-1) - Cognee 1.5.4 API, MCP over Streamable HTTP, and Postgres + pgvector
- [Hermes Agent with Web GUI](https://railway.com/deploy/hermes-agent-with-web-gui) - Deploy and Host Hermes Agent with Web GUI with Railway
- [Tabby | Open-Source AI Code Assistant on Railway](https://railway.com/deploy/tabby) - Self Host Tabby. AI code completions, chat, and codebase search
- [pixelle-video](https://railway.com/deploy/pixelle-video) - AI short-video studio with web UI and flexible model pipelines
- [Paperclip Railway Template](https://railway.com/deploy/paperclip-complete-setup) - Hire AI employees, automate jobs & run your business on autopilot
- [qwenpaw](https://railway.com/deploy/qwenpaw) - Personal AI assistant with memory, skills, and multi-channel support
- [Ferro Labs AI - PostgreSQL storage](https://railway.com/deploy/ferro-labs-ai-postgresql-storage) - Unified AI Gateway - All your traffic. One gateway - PostgreSQL storage.
- [moltis](https://railway.com/deploy/moltis-1) - A Rust-native claw. Voice, memory, MCP tools built-in.
- [mission-control](https://railway.com/deploy/mission-control) - Mission Control is an AI agent orchestration dashboard with task management
- [RAGFlow](https://railway.com/deploy/ragflow) - RAG engine with document parsing, citations, and agentic workflows
- [VoidLLM](https://railway.com/deploy/wild-pure) - Lightweight, privacy-first LLM proxy
- [IronClaw](https://railway.com/deploy/ironclaw) - Deploy IronClaw on Railway with PostgreSQL and pgvector.
- [contextmate](https://railway.com/deploy/contextmate) - Zero-knowledge encrypted sync server for AI agent context
- [MLflow :v3.10.1-full](https://railway.com/deploy/mlflow-full) - MLflow Full version see more: www.oploy.eu
- [free-open-webui-terminal](https://railway.com/deploy/free-open-webui-te-1) - No need for OpenClaw! Your own computer on the cloud, free via Nvidia NIM
- [Hive Agent](https://railway.com/deploy/hive-agent) - A deployment template of hive https://github.com/aden-hive/hive
- [nanobot](https://railway.com/deploy/nanobot-2) - 🐈 nanobot: The Ultra-Lightweight OpenClaw
- [Spacebot](https://railway.com/deploy/spacebot) - An AI agent for teams, communities, and multi-user environments.
- [ragent](https://railway.com/deploy/ragent) - Claude Code in a Docker container. Access from any browser, anywhere.
- [AgentIM](https://railway.com/deploy/agentim) - A unified IM-style platform for AI coding agents
- [OpenClaw – Self-Hosted AI Agent](https://railway.com/deploy/openclaw-prev-clawdbot-moltbot-easy-self) - Self-hosted OpenClaw: deploy your private AI agent on Railway in one click.
- [JupyterLab | Use Open Source Models in Notebooks Directly](https://railway.com/deploy/ollama-jupyter-lab-or-run-local-llms-fro) - [Jun '26] Host any LLM Model using Ollama & Use in Jupyter Notebooks
- [Agentfield Deep Research](https://railway.com/deploy/agentfield-deep-research) - AI powered multi-agent deep research by Agentfield on Railway
- [Agentfield Control Plane](https://railway.com/deploy/controlplane-template) - Kubernetes-style orchestration for multi-agent AI systems with IAM
- [LibreTranslate](https://railway.com/deploy/libretranslate-1) - One-Click deployment of Free and Open Source Machine Translation API
- [LLM Council](https://railway.com/deploy/llm-council) - Multi-model AI deliberation for higher-quality decisions
- [Flowise with n8n](https://railway.com/deploy/flowise-and-n8n-ai-automation) - Automate AI agents, workflows, and integrations in one stack.
- [I Not Moving Today](https://railway.com/deploy/inotmovingtoday) - AI video generator for Reels, TikTok,  Shorts & YouTube
- [GPT-OSS 120B | Self-host GPT OSS 120B on Railway with a chat UI](https://railway.com/deploy/gpt-oss-120b) - [Jun '26] Ensure To Have >64GB Memory Available to Host Successfully
- [GPT-OSS 20B](https://railway.com/deploy/gpt-oss-20b) - [Jun '26] Self-host GPT-OSS 20B on Railway with a chat UI.
- [OpenAI-Agent-SDK](https://railway.com/deploy/openai-agent-sdk) - Deploy and Host OpenAI-Agent-SDK with Railway
- [Nomic Embed](https://railway.com/deploy/nomic-embed) - Generate high-quality text embeddings with Google's best model
- [GPT-OSS](https://railway.com/deploy/gpt-oss) - Deploy and host GPT-OSS on Railway.
- [AI Agentic Automation Stack (n8n)](https://railway.com/deploy/ai-agentic-automation-stack-n8n) - All-in-One AI Automation Stack: workflows, agents, vector DB, web scraping
- [Latitude](https://railway.com/deploy/latitude-1) - Self-hosted AI agent observability with OTLP, evals, and MCP
- [one-api](https://railway.com/deploy/X32mcN) - Access all LLM through the standard OpenAI API format, easy to deploy & use
- [Botpress](https://railway.com/deploy/yEID2r) - Botpress is the ultimate platform for building next-generation chatbots.
- [hugging-face-transformer](https://railway.com/deploy/z52Exi) - A Hugging Face Transformer Railway Template
- [Qdrant Cluster](https://railway.com/deploy/dbhF-C) - High-performance Vector Database and Vector Search Engine
- [Langflow | Open Source Visual Builder for LangChain AI Workflows](https://railway.com/deploy/-spvZa) - Low-code visual builder for RAG and multi-agent AI apps
- [Director by VideoDB](https://railway.com/deploy/QJbo7o) - Director is an open source framework for creating AI agents.
- [LobeChat | Open Source ChatGPT Alternative for Any Model Provider](https://railway.com/deploy/XJrbIL) - Multi-provider AI chat with built-in accounts and a document knowledge base
- [Open WebUI with Pipelines](https://railway.com/deploy/open-webui-with-pipelines) - Open WebUI + Pipelines is an extendable, user friendly self-hosted LLM UI.
- [Anthropic x FastAPI](https://railway.com/deploy/5o0WPC) - Starter for your backend AI-powered app
- [openai-proxy](https://railway.com/deploy/YTHiYS) - OpenAI Proxy (100+ LLMs) - OpenAI, Azure, Bedrock, Anthropic, HuggingFace
- [Chroma](https://railway.com/deploy/tifygm) - the AI-native open-source embedding database
- [Dialoqbase](https://railway.com/deploy/dialoqbase) - Create chatbots with ease
- [Chatbot UI](https://railway.com/deploy/dSKMsy) - An open-source ChatGPT UI alternative.
- [LlamaIndex Apps](https://railway.com/deploy/GpZ0J4) - Sample LlamaIndex apps for chatting with PDFs and URL summaries.
- [perplexity-ai](https://railway.com/deploy/perplexity-ai) - 用你的 Perplexity 订阅额度，自建 OpenAI 兼容 API + MCP 服务，自带号池管理面板与中英文界面。
- [Mastra workflow evaluation demo](https://railway.com/deploy/mastra-workflow-evaluation-demo) - Deterministic Mastra memory and human approval workflow demo.
- [LibreTranslate | (Just Updated) Translation API, Key-Locked, Models Survive Redeploys](https://railway.com/deploy/libretranslate-or-just-updated-translati) - LibreTranslate translation API, key-locked. Models survive redeploys
- [Rustic](https://railway.com/deploy/rustic) - Self-hosted agentic code editor you can use from any browser or phone
- [JevOS](https://railway.com/deploy/jevos) - Offline customer-support triage classifier (local Jev model server)
- [Paperclip](https://railway.com/deploy/paperclip-agent-orchestration) - Paperclip [Oct'26] — run Claude Code and Codex agent teams on Postgres
- [Mobius](https://railway.com/deploy/mobius-ai-agent) - Mobius [Oct'26] — an AI agent that builds apps, with memory that persists
- [OmniRoute](https://railway.com/deploy/omniroute-llm-gateway) - OmniRoute [Oct'26] — one OpenAI-compatible endpoint, your provider keys
- [OpenDots](https://railway.com/deploy/opendots-3) - Self-host AI coworkers with chat, pages, calls, and Slack
- [anythingmcp](https://railway.com/deploy/anythingmcp) - Turn any REST, SOAP, GraphQL, OData or SQL API into MCP tools
- [LibreChat | Stable Release, Working Schedules, Uploads in a Bucket](https://railway.com/deploy/librechat-or-stable-release-working-sche) - Self-host LibreChat on Railway — stable release, uploads in a Bucket.
- [Rove](https://railway.com/deploy/rove) - Early-release AI agent with web chat, MCP tools, and GitHub AIPs.
- [LlamaIndex durable agents](https://railway.com/deploy/llamaindex-durable-agents) - LlamaIndex approvals with DBOS recovery and private PostgreSQL.
- [Factory on Rails](https://railway.com/deploy/factory-on-rails) - Self-hosted coding agents that turn tasks into pull requests
- [OpenDots](https://railway.com/deploy/opendots) - OpenDots by CopilotKit: AI coworkers, pages, and private browsing
- [OpenWeb UI](https://railway.com/deploy/openweb-ui) - Self-host Open WebUI [Oct'26]— private ChatGPT UI with RAG over documents
- [Tallylamp Worker](https://railway.com/deploy/tallylamp-worker) - A worker for Tallylamp: more browsers, on a service with its own limit.
- [magpie](https://railway.com/deploy/magpie-1) - Deploy magpie with web UI on Railway.
- [Codewhale | Cloud Coding Agent (Web Terminal + SSH)](https://railway.com/deploy/codewhale) - Codewhale coding agent, git, gh, Node, Python. Browser terminal + SSH.
- [RSSBox | Feed Translator With Working Redis and a Private Admin](https://railway.com/deploy/rssbox-or-feed-tra-1) - RSSBox on Railway: translated feeds, a Redis that pulls, private admin
- [Dify | AI App Builder with a Redis That Requires a Password](https://railway.com/deploy/dify-or-ai-app-bui-1) - AI app builder stack whose Redis actually enforces its password
- [Lobe Chat | Multi-Provider AI Chat, Pinned and Access-Protected](https://railway.com/deploy/lobe-chat-or-multi-1) - Lobe Chat on Railway: pinned release, access code set, no key prompts
- [Speech | Whisper and Kokoro on One Endpoint, Models on a Volume](https://railway.com/deploy/speech-or-whisper--1) - Speech to text and text to speech on one endpoint, models kept on a volume.
- [LiteLLM Gateway | Virtual Keys and Budgets That Actually Hold](https://railway.com/deploy/litellm-gateway-or-1) - Self-host LiteLLM with virtual keys, budgets and limits that actually hold.
- [LlamaIndex Apps | Streamlit on the Port the Platform Assigns](https://railway.com/deploy/llamaindex-apps-or-1) - Streamlit apps that bind the assigned port. The stock one never does.
- [NextChat | Pinned Release with an Access Code, No Key Required](https://railway.com/deploy/nextchat-or-pinned-1) - NextChat pinned to a release, access code generated, no key demanded
- [Flowise | Visual AI Agent Builder with Postgres and a Real Volume](https://railway.com/deploy/flowise-or-visual--1) - Self-host Flowise on Railway: agents, Postgres and flows that survive
- [Document Search | Upload PDFs, Search Them by Meaning](https://railway.com/deploy/document-search-or-1) - Upload PDFs and search them by meaning: Docling, embeddings, pgvector.
- [Langflow | Visual AI Workflow Builder with Postgres and a Real Volume](https://railway.com/deploy/langflow-or-visual-1) - Self-host Langflow on Railway — flows, Postgres & files that persist.
- [Bifrost | Go AI Gateway That Is Closed by Default](https://railway.com/deploy/bifrost-or-go-ai-g-1) - Self-host the Bifrost AI gateway — admin auth on, /v1 closed by default.
- [MCP Server Starter | Remote Streamable HTTP with OAuth 2.1 & Postgres](https://railway.com/deploy/mcp-server-starter-1) - Host your own remote MCP server — Streamable HTTP, OAuth 2.1, Postgres.
- [Ownstate](https://railway.com/deploy/ownstate) - Self-hosted institutional memory and context for secure AI agents.
- [Arcnem Vision](https://railway.com/deploy/arcnem-vision) - Image analysis and agent workflows with OCR, search, and segmentation.
- [anydoc | (Just Updated) Any Document to Markdown API, Japanese & Thai OCR](https://railway.com/deploy/anydoc-or-just-updated-any-document-to-m) - Word, PowerPoint, Excel, PDF and scans to Markdown; Japanese/Thai OCR
- [PanWatch](https://railway.com/deploy/panwatch) - AI stock monitoring dashboard with alerts and paper trading
- [opencode-telegram-bot](https://railway.com/deploy/opencode-telegram-bot) - Telegram client for OpenCode coding tasks
- [Open Notebook](https://railway.com/deploy/open-notebook-2) - Open Notebook 1.14: self-hosted NotebookLM alternative with SurrealDB.
- [MCPHub](https://railway.com/deploy/mcphub) - MCPHub 1.0: one endpoint and dashboard for all your MCP servers.
- [Nanobot](https://railway.com/deploy/nanobot-1) - nanobot 0.3.5: lightweight personal AI agent with a web UI and chat apps.
- [lightpanda](https://railway.com/deploy/lightpanda-1) - Lightpanda 0.4: fast headless browser for agents over CDP, token-gated.
- [Nimbus Agent - 24/7 Employee](https://railway.com/deploy/nimbus-agent-247-employee) - Deploy a fully autonomous 24/7 AI employee with built-in MCP tools.
- [MarkItDown](https://railway.com/deploy/markitdown) - Microsoft MarkItDown 0.1 document-to-Markdown REST and MCP service.
- [ContextForge](https://railway.com/deploy/contextforge-1) - IBM ContextForge 1.0 MCP gateway and registry with auth, on Postgres.
- [Bifrost](https://railway.com/deploy/bifrost-2) - Bifrost 2.2 fast AI gateway for 20+ LLM providers with virtual keys.
- [Label Studio](https://railway.com/deploy/label-studio-2) - Label Studio 1.23 data labeling for text, images and audio, on Postgres.
- [open-glean](https://railway.com/deploy/open-glean) - Self-hosted AI knowledge workspace with Hydra DB search and citations
- [Text Embeddings Inference](https://railway.com/deploy/text-embeddings-in-1) - Hugging Face Text Embeddings Inference 1.9 on CPU with an API key.
- [Mem0](https://railway.com/deploy/mem0-1) - Mem0 2.2 self-hosted AI memory API and dashboard with Postgres pgvector.
- [Livekit [Updated Oct'26]](https://railway.com/deploy/livekit-agent-worker) - Run a LiveKit voice agent worker — STT, LLM, TTS, sub-second replies
- [YouTube Transcript MCP](https://railway.com/deploy/youtube-transcript-mcp) - Connect YouTube to Claude and ChatGPT with cited transcripts
- [omnigent-server](https://railway.com/deploy/omnigent-server) - Self-hosted control plane for AI coding agents; Postgres + admin seeded
- [memoh](https://railway.com/deploy/memoh) - Every agent gets its own computer, desktop, network, and long-term memory
- [harnessrouter](https://railway.com/deploy/harnessrouter-1) - Unified interface for agent harnesses.
- [harnessrouter](https://railway.com/deploy/harnessrouter) - Self-hosted AI agent workspace with web UI
- [Speakr](https://railway.com/deploy/speakr) - Self-hosted AI audio transcription & notes with your own AI providers
- [MCP Atlassian](https://railway.com/deploy/mcp-atlassian) - A remote MCP server for Jira & Confluence (streamable-HTTP, multi-user).
- [GitHub Stars Manager](https://railway.com/deploy/github-stars-manager) - AI semantic search & auto-categorization for your GitHub stars.
- [AnyCrawl](https://railway.com/deploy/anycrawl) - Self-hosted scrape/crawl/SERP API that turns sites into LLM-ready data.
- [MemPalace](https://railway.com/deploy/mempalace) - Self-hosted AI memory server (MCP) with a private Qdrant, for agents.
- [Local Deep Research](https://railway.com/deploy/local-deep-research) - AI deep-research assistant with citations and private SearXNG, behind auth
- [DeepWiki-Open](https://railway.com/deploy/deepwiki-open) - AI-generated interactive wikis for any Git repo, behind a password
- [Daily Stock Analysis](https://railway.com/deploy/daily-stock-analysis) - AI daily reports over your stock watchlist, with a secured dashboard
- [Marmot](https://railway.com/deploy/marmot) - The open-source context layer for AI agents.
- [prompts.chat](https://railway.com/deploy/promptschat) - Self-hosted AI prompt library: 2,000 CC0 prompts, MCP, closed sign-up
- [InsightsLM](https://railway.com/deploy/insightslm) - Open-source NotebookLM: chat with sources, audio overviews; Supabase + n8n
- [LLM Wiki](https://railway.com/deploy/llm-wiki) - Karpathy's LLM wiki, self-hosted: Claude builds your wiki over MCP
- [ReMe](https://railway.com/deploy/reme) - Self-evolving Markdown memory your AI agents share over MCP
- [Mike Legal AI](https://railway.com/deploy/mike-legal-ai) - Open-source legal AI for document review and drafting, fully self-hosted
- [Morphic + SearXNG](https://railway.com/deploy/morphic-searxng) - Perplexity-style AI answer engine with accounts and private SearXNG search
- [Kortix + Supabase](https://railway.com/deploy/kortix-supabase) - AI agent platform with cloud sandboxes and a self-hosted Supabase
- [PenguinHarness](https://railway.com/deploy/penguinharness) - Multi-agent platform where agents build, evaluate and optimize agent apps
- [Coddy](https://railway.com/deploy/coddy) - AI agent with a web UI, OpenAI-compatible API, scheduler and remote mode
- [Sirene Inference](https://railway.com/deploy/sirene-inference) - Speech generation worker for Sirene, self-registering on first start.
- [LangWatch | Open Source LangSmith Alternative for Agent Testing](https://railway.com/deploy/langwatch-llm-observability) - Self-hosted LangWatch: LLM tracing, evals, guardrails and agent testing
- [Yao Agents](https://railway.com/deploy/yao-agents) - Self-hosted AI agent hub: workspaces, task board, Open API and mobile apps
- [Mirage Daemon](https://railway.com/deploy/mirage-daemon) - Mirage daemon — a versioned virtual filesystem and shell for AI agents
- [Ubuntu AI Coding Agents Terminal (Web + SSH)](https://railway.com/deploy/ubuntu-ai-coding-agents-terminal-web-ssh) - Web terminal + SSH: Claude Code, Codex, Gemini, Copilot, OpenCode, aider
- [tlbx](https://railway.com/deploy/tlbx) - Browser terminal and coding-agent workstation with a persistent volume
- [BankMCP](https://railway.com/deploy/bankmcptm) - Deploy and Host BankMCP™ on Railway
- [OpenKnowledge](https://railway.com/deploy/openknowledge) - Self-hosted knowledge base for AI agents, with MCP and Google sign-in
- [Firecrawl Full Stack](https://railway.com/deploy/firecrawl-full-stack) - Firecrawl web scraping API with Playwright, Redis, RabbitMQ, and Postgres
- [Open Notebook + SurrealDB](https://railway.com/deploy/open-notebook-surrealdb) - Self-hosted NotebookLM alternative with SurrealDB, notes, chat and podcasts
- [cortex-engine](https://railway.com/deploy/cortex-engine) - Persistent memory for AI agents: belief graph, REST API, your own LLM key
- [MCPJungle Enterprise Mode](https://railway.com/deploy/mcpjungle-enterprise-mode) - MCP registry and gateway with PostgreSQL and a protected initialization.
- [DB-GPT Data Assistant](https://railway.com/deploy/db-gpt-data-assistant) - Private AI data assistant with persistent metadata and vector storage.
- [Prompt Optimizer | Open Source PromptPerfect Alternative](https://railway.com/deploy/prompt-optimizer) - AI prompt optimizer web app and MCP server, password and token protected
- [paperless-gpt](https://railway.com/deploy/paperless-gpt) - Add AI OCR, titles and tags to your paperless-ngx, behind a password
- [Text Embeddings Inference](https://railway.com/deploy/text-embeddings-inference) - Self-hosted embeddings and reranking API, behind an API key
- [kotaemon](https://railway.com/deploy/kotaemon) - Self-hosted RAG chat over your documents, with a real admin password
- [Whisper ASR Webservice](https://railway.com/deploy/whisper-asr-webservice) - Self-hosted Whisper speech-to-text API with subtitles, behind a password
- [LongMemory](https://railway.com/deploy/longmemory) - Persistent memory for LLM apps and agents, with MCP and a dashboard
- [Background Removal API | (Just Updated) rembg, Weights Baked In, 2GB RAM Floor](https://railway.com/deploy/background-removal-api-or-just-updated-r) - rembg cutouts, weights baked in, commercial-safe models only, 2GB RAM floor
- [nolock + llama.cpp: ai coding harness + local slms (micro-agents)](https://railway.com/deploy/nolock-llamacpp-ai-coding-harness-local-) - ai coding harness + local slms (micro-agents)
- [ArcticWorks CodePeer](https://railway.com/deploy/arcticworks-codepeer) - Self-hosted AI code reviewer for GitHub pull requests.
- [OpenPencil](https://railway.com/deploy/openpencil) - Private AI vector design workspace with password protection and storage.
- [EverOS](https://railway.com/deploy/everos) - EverOS agent memory server behind a bearer-token gateway, data on a volume
- [LibreChat [Updated Sep '26]](https://railway.com/deploy/librechat-open-source) - LibreChat [Sep '26] (Self-Hosted ChatGPT UI, Multi-Provider + RAG)
- [Observal](https://railway.com/deploy/observal) - Registry and insight engine for coding-agent Skills, MCP servers, Agents.
- [gortex](https://railway.com/deploy/gortex) - Code-intelligence MCP server for AI agents, with web UI.
- [Firecrawl](https://railway.com/deploy/firecrawl-1) - Web scraping and crawling API with authentication and persistent queues.
- [NOFX AI Trading Terminal](https://railway.com/deploy/nofx-2) - NOFX AI trading dashboard for evaluation; no live trading verified.
- [Langflow](https://railway.com/deploy/langflow-2) - Build AI workflows with Langflow, private Postgres and persistent storage.
- [NextChat](https://railway.com/deploy/nextchat-1) - NextChat with strict access-code protection and a required server-side provider key.
- [projectmem](https://railway.com/deploy/projectmem) - Open-source coding agent memory as a hosted MCP server.
- [Flowise | AI Agents with Workers](https://railway.com/deploy/flowise-workers) - Flowise queue mode: main + worker, Postgres (pgvector), Redis, S3 bucket.
- [LobeChat](https://railway.com/deploy/lobe-chat-railway) - AI chat and agent workspace that uses your own model API keys
- [utopia](https://railway.com/deploy/utopia) - World's first open-source enterprise world model.
- [notal](https://railway.com/deploy/notal) - Git workspace for humans + agents — Markdown queue with receipts, MCP.
- [WhatsApp AI Business Hub](https://railway.com/deploy/whatsapp-ai-business-hub) - WhatsApp AI support, CRM inbox, and automation
- [Portkey](https://railway.com/deploy/portkey) - A powerful AI gateway for routing, guardrails, fallbacks, and 200+ LLMs.
- [Code Reviewer (Mira + MCP)](https://railway.com/deploy/code-reviewer-mira-mcp) - Always-on AI GitHub PR reviewer with dashboard, MCP, and your own LLM keys.
- [Skyvern](https://railway.com/deploy/skyvern-ai) - Self-hosted Skyvern: LLM browser automation, workflows, credential vault
- [LocalAI | (Just Updated) Self-Hosted OpenAI API With The Model Baked In](https://railway.com/deploy/localai-or-just-updated-self-hosted-open) - OpenAI-compatible LLM API. Model and backend baked in, ready on first boot.
- [trueforge](https://railway.com/deploy/trueforge-1) - Hosted TrueForge: agent harness with UI, Postgres & Redis
- [dify 1.17](https://railway.com/deploy/dify-117) - Deploy Dify 1.17 on Railway — LLM apps, RAG, agents, plugins, and workflows
- [Archon](https://railway.com/deploy/archon) - Run Claude Code from a browser, behind real per-user login.
- [Screenshot to Code](https://railway.com/deploy/shotcode-source) - Turn screenshots and Figma designs into editable frontend code with AI.
- [Chroma](https://railway.com/deploy/chroma-db) - Persistent vector database for RAG and AI memory with protected access.
- [Milvus](https://railway.com/deploy/milvus) - Scalable vector database for GenAI applications and semantic search.
- [CLIProxyAPI — Daily Auto-Update](https://railway.com/deploy/cliproxyapi-secure-release-tracked) - CLIProxyAPI gateway with verified daily updates and automatic rollback.
- [Hindsight — AI Agent Memory](https://railway.com/deploy/hindsight-ai-agent-memory) - Memory for AI agents that learns, recalls, and evolves over time.
- [AI Gateway + Observability (LiteLLM × Langfuse)](https://railway.com/deploy/ai-gateway-observability-litellm-langfus) - [Sep'26] LiteLLM gateway + Langfuse tracing, pre-wired. 100+ LLMs
- [Claude Agent SDK Worker](https://railway.com/deploy/claude-agent-sdk-worker) - [Sep'26] Scheduled autonomous Claude agents — official SDK + run history
- [Cloudflare OS Operator](https://railway.com/deploy/cloudflare-os-operator) - Deploy and upgrade Cloudflare OS from a secure Railway operator console.
- [nasiko](https://railway.com/deploy/nasiko) - AI agent control plane with embedded dashboard, routing, and observability.
- [nakama](https://railway.com/deploy/nakama) - It's like Hermes Agent & OpenClaw but designed to work nicely with teams.
- [Langflow | (Just Updated) Visual AI Agent Builder That Actually Boots](https://railway.com/deploy/langflow-or-just-updated-visual-ai-agent) - Boots first try: admin seeded, files kept, DB private, version pinned.
- [Mobius — Personal AI OS](https://railway.com/deploy/mobius-personal-ai-os) - Your self-hosted AI agent that builds apps, learns, and gets things done.
- [Open Wearables | (Just Updated) Wearable Health API, Admin Password Not admin@admin.com](https://railway.com/deploy/open-wearables-or-just-updated-wearable-) - Wearable health API with a sealed admin account, not the published default
- [PI WEB | (Just Updated) Pi Coding Agent In Your Browser, Password Protected](https://railway.com/deploy/pi-web-or-just-updated-pi-coding-agent-i) - Pi Coding Agent web UI: basic-auth enforced, workspaces survive redeploys
- [Firecrawl | (Just Updated) Web Scraper API for LLMs, Not Open to the Internet](https://railway.com/deploy/firecrawl-or-just-updated-web-scraper-ap) - API-key-gated web scraper API for LLMs. Queue, cache and renderer wired.
- [Letta | (Just Updated) Stateful Agent Server Whose Memory Survives Redeploys](https://railway.com/deploy/letta-or-just-updated-stateful-agent-ser) - Agent memory that survives redeploys, on a password-protected agent API.
- [AutoMem | (Just Updated) Agent Memory That Actually Embeds](https://railway.com/deploy/automem-or-just-updated-agent-memory-tha) - Agent memory that really embeds: model baked in, graph password enforced
- [Presenton](https://railway.com/deploy/presenton-ai-slides) - Turns a prompt or document into an editable PowerPoint presentation
- [Horizon](https://railway.com/deploy/horizon-aug-26-ai-news-radar) - AI reads your news sources and writes a daily briefing
- [Edict [Aug '26] (Multi-Agent Dashboard)](https://railway.com/deploy/edict-aug-26-multi-agent-dashboard) - Nine AI agents you monitor from one live dashboard
- [DEEIX-Chat](https://railway.com/deploy/calm-wide) - Route prompts to the right AI model from one dashboard
- [LiteLLM (Self-Hosted AI Gateway)](https://railway.com/deploy/litellm-self-hosted-ai-gateway) - Self-hosted LLM proxy for any model, with keys, budgets, and usage logs.
- [Langfuse v4 | Open Source LangSmith Alternative for LLM Observability](https://railway.com/deploy/langfuse-v4) - Self-hosted LLM tracing and evals on Langfuse v4, old and new SDKs
- [OpenHands | (Just Updated) AI Coding Agent Whose API Key Isn't Public](https://railway.com/deploy/openhands-or-just-updated-ai-coding-agen) - Current Agent Canvas 1.14.0. Session key never baked into the page.
- [vast-wild](https://railway.com/deploy/vast-wild) - What is vast-wild? Your descrip
- [Ollama + Open WebUI — Private AI Chat Stack](https://railway.com/deploy/ollama-openwebui-chat) - Self-host a private AI chat UI with local & cloud models
- [MCP JS](https://railway.com/deploy/mcp-js) - mcp js gives a snapshottable javascript repl connected to your tools for ai
- [Tasuku](https://railway.com/deploy/tasuku) - Self-hosted AI agent control plane for GitHub and Slack
- [OpenClaw on NixOS | (Just Updated) AI Agent That Installs Its Own Tools](https://railway.com/deploy/openclaw-on-nixos-or-just-updated-ai-age) - Self-hosted AI agent on nix. Installs any tool at runtime. 1 GB RAM min.
- [Banana Slides](https://railway.com/deploy/bananaslides) - AI slide maker: turn prompts or files into editable PPTs with voice.
- [Grep Knowledge Agent](https://railway.com/deploy/grep-knowledge-agent) - Self-hosted AI knowledge agent — grep, not vector embeddings
- [OpenHands — Autonomous GitHub Issue Resolver](https://railway.com/deploy/openhands-issue-resolver) - Self-host OpenHands — assign a GitHub issue, get a pull request
- [Open Notebook | Open Source NotebookLM Alternative](https://railway.com/deploy/open-notebook-notebooklm) - Self-hosted NotebookLM alternative - chat with docs, notes, podcasts
- [Khoj — Self-Hosted AI Second Brain](https://railway.com/deploy/khoj-ai-second-brain) - Self-host Khoj — chat with your docs, AI second brain
- [R2R — Self-Hosted Agentic RAG Engine](https://railway.com/deploy/r2r-rag-engine) - Self-host R2R — agentic RAG, hybrid search & knowledge graphs
- [Promptfoo — Self-Hosted LLM Eval Dashboard](https://railway.com/deploy/promptfoo-llm-evals) - Self-host Promptfoo — private dashboard for LLM eval results
- [Xinference — Self-Hosted Embeddings, Rerankers & LLMs](https://railway.com/deploy/xinference-inference-api) - Self-host an OpenAI-compatible endpoint — embeddings & rerankers
- [ElizaOS — Self-Hosted AI Agent with Memory](https://railway.com/deploy/elizaos-ai-agent) - Self-host an ElizaOS agent — character AI with pgvector memory
- [nanobot — Reliable AI Assistant](https://railway.com/deploy/nanobot-reliable-ai-assistant) - Community nanobot 0.2.2: pinned, token-gated, persistent, healthchecked.
- [CoPaw - AI Assistant](https://railway.com/deploy/copaw-ai) - Versatile local/cloud AI assistant with memory, tasks, and chat channels.
- [Qdrant | Open Source Vector Database for AI and RAG](https://railway.com/deploy/qdrant-vector-store) - Vector database for AI search and RAG — API key enabled, data on a volume
- [ElizaOS](https://railway.com/deploy/elizaos-secure-starter) - Authenticated ElizaOS agent with OpenAI and persistent pgvector memory
- [Cursor Agent Machine](https://railway.com/deploy/cursor-agent-machine) - Cursor Agent Machine hosted on railway
- [Mastra Agent on Oracle Database](https://railway.com/deploy/mastra-agent-on-oracle-database) - Mastra weather agent with all state in Oracle Database Free.
- [Mastra Voice Agent with LiveKit](https://railway.com/deploy/mastra-voice-agent-with-livekit) - Call-center voice agent on LiveKit with Mastra memory and replies.
- [Mastra Agent Builder](https://railway.com/deploy/mastra-agent-builder) - Hosted Studio agent builder with MastraEditor. Enterprise license needed.
- [Mastra Agent Showcase](https://railway.com/deploy/mastra-agent-showcase) - The Mastra kitchen sink: agents, workflows, MCP, guardrails, HITL.
- [Mastra Evals with Memory](https://railway.com/deploy/mastra-evals-with-memory) - Run Mastra evals on memory-enabled agents. Mock model, no API keys.
- [OpenHuman Core | (Just Updated) Remote AI Core, Healthchecked](https://railway.com/deploy/openhuman-core-v06-1) - Remote AI core with a working healthcheck and the Node runtime prebaked
- [VoicePro](https://railway.com/deploy/voicepro) - AI toolkit for voice cloning, TTS, and translation.
- [Odysseus | (Just Updated) PewDiePie AI Workspace Whose Web Search Works](https://railway.com/deploy/odysseus-v102-or-pewdiepie-ai-workspace-) - AI workspace whose web search returns results instead of 403
- [Agentmemory | (Just Updated) Agent Memory Whose Recall Stays Scoped](https://railway.com/deploy/agentmemory-v0928-or-agent-memory-whose-) - Agent memory service whose recall stays scoped to the agent that wrote it
- [MinerU | (Just Updated) PDF-to-Markdown API for RAG, CPU-Ready and Preloaded](https://railway.com/deploy/mineru-v344-or-pdf-to-markdown-api-for-r) - PDF-to-Markdown API for RAG, CPU-ready with models already preloaded
- [Presenton | (Just Updated) AI Presentations + API, Decks Survive Redeploy](https://railway.com/deploy/presenton-v093-or-ai-presentations-api-d) - AI slide decks plus an API; presentations survive every redeploy
- [Qdrant — High-Performance Vector Database](https://railway.com/deploy/qdrant-vector-search) - Self-host Qdrant — fast vector search at scale for RAG & AI
- [Chroma — Self-Hosted AI Vector Database](https://railway.com/deploy/chroma-vector-database) - Self-host Chroma — vector database for RAG & semantic search
- [Langfuse — Self-Hosted LLM Observability](https://railway.com/deploy/langfuse-llm-observability) - Self-host Langfuse — LLM traces, evals & prompt management
- [SearXNG | (Just Updated) AI Agent Search API, 16x Concurrent Searches](https://railway.com/deploy/searxng-v202683-or-ai-agent-search-api-1) - Search API for agents: JSON results answer on the first deploy, not 403
- [Presidio](https://railway.com/deploy/presidio-pii) - Authenticated PII detection and anonymization API
- [magic-resume](https://railway.com/deploy/magic-resume-1) - AI-powered resume editor with live preview and PDF export
- [OneHub - Updated One API](https://railway.com/deploy/onehub) - Access all LLMs through the standard OpenAI API format,Analytics & billing.
- [nerve-template-source](https://railway.com/deploy/nerve-template-source) - Visual command center for securely operating OpenClaw and Hermes agents.
- [Flowise | (Just Updated) Visual AI Agent Builder, Credentials Survive Redeploys](https://railway.com/deploy/flowise-v314-or-visual-ai-agent-builder-) - Visual AI agent builder whose provider credentials survive a redeploy
- [Agentmemory — Shared Memory for AI Coding Agents](https://railway.com/deploy/agentmemory-coding-agent-memory) - Self-host shared memory for Claude Code, Cursor & Codex agents
- [2Apifare](https://railway.com/deploy/2apifare) - GeminiCLI into OpenAI,GEMINI API, supporting Antigravity models.
- [Dify2Openai](https://railway.com/deploy/dify2openai) - Convert Dify API into an OpenAI API,streaming,blocking,chat,agent & more
- [Paperclip | AI Agent Company OS, Current Release + No Build Step](https://railway.com/deploy/paperclip-or-ai-agent-company-os-current) - Run a company with AI agents. Current release, no build step, empty form.
- [Jig](https://railway.com/deploy/jig) - Reliable AI workflows with versioned TypeScript and scoped MCP tools.
- [x402 Commerce](https://railway.com/deploy/x402-commerce) - Agent-to-Agent Micropayment Infrastructure with x402 v2 and MCP server
- [JimiHub - Gemini to OpenAI Proxy](https://railway.com/deploy/jimihub) - Centralized Gemini API proxy, key management, and quota tracker.
- [Open WebUI](https://railway.com/deploy/open-webui-7) - Self-hosted AI chat for OpenAI-compatible APIs and local models.
- [AnythingLLM](https://railway.com/deploy/anythingllm) - Private AI workspace for chat, documents, agents, and RAG.
- [AI Short Video Generator — Gemini + Veo](https://railway.com/deploy/ai-short-video-generator-veo) - Generate TikTok, Reels & Shorts videos with AI — Gemini + Veo
- [LobeChat [Jul'26]](https://railway.com/deploy/lobechat-jul26) - Self-hosted AI chat with memory and plugins
- [InvokeAI — Self-Hosted Stable Diffusion Studio](https://railway.com/deploy/invokeai-stable-diffusion-studio) - Self-host InvokeAI — canvas, node workflows & custom SD models
- [OpenHuman Core — Self-Hosted AI Memory Backend](https://railway.com/deploy/openhuman-core-rpc-backend) - Host OpenHuman's core & memory 24/7 — RPC backend for the app
- [Apify MCP Server](https://railway.com/deploy/apify-mcp-server-1) - Apify MCP Server — web scraping and data extraction tools for AI agents
- [Ollama](https://railway.com/deploy/ollama) - Run open-source LLMs with Ollama and persistent model storage.
- [LibreTranslate](https://railway.com/deploy/libretranslate-2) - Self-hosted machine translation API with persistent API-key storage.
- [Open WebUI](https://railway.com/deploy/open-webui-5) - Self-hosted AI chat interface for Ollama and OpenAI-compatible models.
- [banana-slides](https://railway.com/deploy/banana-slides) - AI-native slide generation with templates, editing, and PPTX export
- [Multica — Self-Hosted AI Agent Teammates](https://railway.com/deploy/multica-agent-platform) - Assign tasks to Claude Code & Codex agents — self-hosted platform
- [LightRAG — Knowledge Graph RAG [Postgres Production]](https://railway.com/deploy/lightrag-postgres-graph-rag) - Knowledge-graph RAG on Postgres — Web UI, 20+ LLM providers
- [Nanobot — Multi-LLM Gateway with Dashboard](https://railway.com/deploy/nanobot-llm-gateway) - Route OpenAI, Gemini & Groq with fallback — Telegram-ready gateway
- [Label Studio — Open Source Data Labeling [Postgres]](https://railway.com/deploy/label-studio-data-labeling-postgres) - Self-host Label Studio — label images, text, audio, video
- [Toolkit](https://railway.com/deploy/toolkit) - Process image, video, audio & documents via local AI (Snapotter )
- [FlowSpeech Railway Template](https://railway.com/deploy/flowspeech-railway-template) - Deploy a minimal FlowSpeech text to speech AI starter.
- [Engram](https://railway.com/deploy/engram) - Git-backed markdown memory for AI agents that knows what's still true
- [deltallm](https://railway.com/deploy/deltallm) - Deploy DeltaLLM with managed PostgreSQL and Redis on Railway.
- [hippocamp](https://railway.com/deploy/hippocamp) - Run Hippocamp Dream overnight and open reviewable memory compaction PRs.
- [Sim — Self-Hosted Visual AI Agent Builder](https://railway.com/deploy/sim-ai-agent-builder) - Self-host Sim: visual AI agent builder. 1,000+ integrations, BYO LLM.
- [Qdrant — Self-Hosted Vector Database & Pinecone Alternative](https://railway.com/deploy/qdrant-vector-db) - Self-host Qdrant: fast vector database for RAG. No per-query Pinecone fees.
- [Gemini Balance](https://railway.com/deploy/gemini-balance) - Gemini API Proxy and Load Balancer
- [frontman](https://railway.com/deploy/frontman) - Self-host Frontman, browser-native AI frontend agent. frontman.sh
- [GPT Researcher](https://railway.com/deploy/gpt-researcher) - Autonomous agent that conducts deep research on data using LLM providers
- [Devin AI — Telegram Bot](https://railway.com/deploy/devin-ai-telegram-bot) - Deploy Devin, your friendly AI software engineer, directly into Telegram.
- [Web UI - kie.ai](https://railway.com/deploy/web-ui-kie-ai) - Deploy and Host KIE AI Image Generator with Railway
- [outpost](https://railway.com/deploy/outpost) - AI coding agent with webhook automation and built-in dashboard
- [Graphiti + FalkorDB Knowledge Graph Memory](https://railway.com/deploy/graphiti-falkordb-knowledge-graph-memory) - Real-time temporal knowledge-graph MCP server on low-RAM FalkorDB.
- [LightCrawl with Redis](https://railway.com/deploy/lightcrawl-with-redis) - Scalable LightCrawl with Redis backend for distributed queue-based crawling
- [LightCrawl](https://railway.com/deploy/lightcrawl) - Playwright-based lightweight scraping API and MCP server
- [NextChat on Bifrost: Self Hosted ChatGPT UI for Any Model](https://railway.com/deploy/nextchat-on-bifrost-self-hosted-chatgpt-)
- [Devin AI — Discord Bot](https://railway.com/deploy/devin-ai-discord-bot) - Deploy Devin, your friendly AI software engineer, directly into Discord.
- [Langflow on Bifrost: Visual AI Agent Builder, Any Model](https://railway.com/deploy/langflow-on-bifrost-visual-ai-agent-buil) - Host Langflow on Bifrost: Visual AI Agent Builder, Any Model with Railway
- [AnythingLLM on Bifrost: Self Hosted Doc Chat and RAG Agents](https://railway.com/deploy/anythingllm-on-bifrost-self-hosted-doc-c) - Deploy and Host AnythingLLM: Self Hosted Doc Chat and RAG Agents
- [Devin Discord Bot](https://railway.com/deploy/devin-discord-bot) - Deploy and Host Devin Discord Bot with Railway
- [owui-litellm-src-20260617b](https://railway.com/deploy/owui-litellm-src-20260617b) - Open WebUI with a private LiteLLM gateway, Postgres state, and Redis cache.
- [litellm-source-20260617](https://railway.com/deploy/litellm-source-20260617) - LiteLLM AI gateway with Postgres keys, Redis cache, and spend controls.
- [SafeAgent Execution Guard](https://railway.com/deploy/elegant-serenity) - Exactly-once guard for AI agents. Prevents duplicate payments on retry.
- [mateclaw](https://railway.com/deploy/mateclaw) - Team-ready self-hosted AI agent workspace with web UI
- [Vane on Bifrost: Self Hosted AI Search Engine](https://railway.com/deploy/vane-on-bifrost-self-hosted-ai-search-en) - Deploy and Host Vane on Bifrost: Self Hosted AI Search Engine with Railway
- [Flowise + Bifrost | Visual AI Agent Builder with a Built In Gateway](https://railway.com/deploy/flowise-bifrost-or-visual-ai-agent-build) - Flowise + Bifrost | Visual AI Agent Builder with a Built In Gateway
- [graphiti-mcp-but-working-railway](https://railway.com/deploy/graphiti-mcp-but-working-railway) - Railway-friendly template/fork of michabbb/graphiti-mcp-but-working
- [n8n+Bifrost| Self Hosted AI Workflow Automation with a Built In Gateway](https://railway.com/deploy/n8nbifrostor-self-hosted-ai-workflow-aut) - n8n + Bifrost| Self Hosted AI Workflow Automation with a Built In Gateway
- [Agentic Data Stack Lite](https://railway.com/deploy/agentic-data-stack-lite) - LibreChat, ClickHouse, Langfuse: an open-source agentic analytics stack
- [NOFX](https://railway.com/deploy/nofx-1) - AI trading terminal for stocks, forex, crypto, and commodities.
- [LibreChat — Multi-Provider AI Chat with RAG](https://railway.com/deploy/librechat-multi-provider-rag) - Self-hosted ChatGPT UI — GPT, Claude, Gemini, RAG, multi-user
- [PDPP Core Template Source](https://railway.com/deploy/pdpp-core-template-source) - Deploy a PDPP Core node with Railway Postgres.
- [Bifrost + PostgreSQL](https://railway.com/deploy/blue-dark) - Deploy Bifrost AI Gateway with PostgreSQL — production-ready in one click.
- [Pewdiepie Odysseus | Personal AI workspace](https://railway.com/deploy/odysseus-or-personal-ai-workspace) - chat, autonomous agents, tools, model serving, email, research, and more
- [pilotdeck](https://railway.com/deploy/pilotdeck) - Open-source AI Agent operating system for task-oriented productivity
- [BabyChain](https://railway.com/deploy/babychain) - Every output becomes the next input.
- [OpenViking : OpenAI [Jun 26'] - Context Database for Hermes, Openclaw, AI Agents](https://railway.com/deploy/openviking-openai-jun-26-context-databas) - Self-Hosted Context Database for Hermes, Openclaw, AI Agents. OpenAI Embeds
- [LangServe RAG API + pgvector](https://railway.com/deploy/langserve-pgvector-rag) - LangServe RAG API with FastAPI, Postgres, pgvector, and auth.
- [RustyRed-GraphDB](https://railway.com/deploy/rustyred-graphdb) - Deploy and Host RustyRed-GraphDB with Railway
- [Railway RAG Stack](https://railway.com/deploy/railway-rag-stack) - Local LLM + pgvector RAG — no external AI APIs needed
- [Everruns](https://railway.com/deploy/everruns) - Deploy the Everruns OSS agent platform on Railway.
- [Atomic](https://railway.com/deploy/atomic-1) - An AI-native knowledge graph that's yours end-to-end.
- [Paperclip - AI Agent Company (Lite)](https://railway.com/deploy/paperclip-ai-agent-company-lite) - Open-source control plane for AI agents. Single service, embedded Postgres.
- [DeltaNode Dev Agents](https://railway.com/deploy/deltanode-dev-agents) - AI agent workflow system with validation scoring and self-improvement loops
- [AutoVault](https://railway.com/deploy/autovault) - A local-first vault for validated, signed, scoped skills used by AI agents.
- [VantagePeers MCP](https://railway.com/deploy/vantagepeers-mcp) - AI agent coordination: memory, messaging, tasks, knowledge. + OAuth 2.1.
- [Flowise with Workers](https://railway.com/deploy/flowise-with-workers) - [Jul'26] Flowise with workers, Redis, Postgres & persistent storage. 🚀
- [Smooth Operator](https://railway.com/deploy/smooth-operator) - an agent to manage your daily ops
- [PageIndex](https://railway.com/deploy/pageindex) - No vector DB, no chunking. LLM reasoning-based RAG with 98.7% accuracy.
- [DeepRead](https://railway.com/deploy/deepread) - AI document processing — OCR, form fill, PII redaction. Free 2k pages/mo.
- [space-agent](https://railway.com/deploy/space-agent) - The agent that re-shapes the Space
- [Plandex | Open Source AI Coding Agent for Large Codebases](https://railway.com/deploy/plandex) - Self Host Plandex. AI coding, plan versioning, multi-file edits, & more
- [replicate-openai](https://railway.com/deploy/replicate-openai) - OpenAI-compatible gateway for Replicate models. BYOK supported.
- [Promptfoo | Open-Source LLM Testing Platform](https://railway.com/deploy/promptfoo-llm-evaluation) - Self host Promptfoo on Railway — evaluate, compare, and red-team LLMs
- [Xinference (Xorbits Inference) | OpenAI-Compatible API](https://railway.com/deploy/xinference) - Self-host Xinference. Self host LLMs, embeddings, rerankers & more.
- [Fabric](https://railway.com/deploy/fabric) - Fabric framework API with bundled AI prompt patterns and persistent config
- [pi-tps-web](https://railway.com/deploy/pi-tps-web) - A web interface to capture pi session customType exports from pi-tps.
- [higress](https://railway.com/deploy/higress) - All-in-one Higress API gateway with instant Railway web console
- [axonhub](https://railway.com/deploy/axonhub) - Unified AI gateway with web UI, routing, and provider compatibility
- [Mercury Agent Railway - 1 Click deploy](https://railway.com/deploy/mercury-agent-railway) - Mercury is a persistent AI agent that runs 24/7.
- [ashim](https://railway.com/deploy/ashim) - Self-hosted AI image studio with 45+ tools, pipelines, and REST API
- [Flowise -  LLM Agent & RAG Builder [Updated Jun'26]](https://railway.com/deploy/flowise-llm-agent-and-rag-builder) - Self-hosted drag-and-drop builder for LLM agents and RAG pipelines.
- [wide-warm](https://railway.com/deploy/wide-warm) - Deploy and Host CrewForm with Railway
- [NakatomiCRM](https://railway.com/deploy/nakatomicrm) - Headless agent-native CRM — REST, MCP, A2A, ACP.
- [Llama Index](https://railway.com/deploy/llama-index) - Llama Index w/Basic Auth (Updated Apr 2026)
- [Multi Agent System](https://railway.com/deploy/multi-agent-system) - Deploy a multi-agent system on Railway, each agent as a service
- [vibe-trading](https://railway.com/deploy/vibe-trading) - AI trading workspace with web UI, backtesting, and multi-agent analysis
- [Rift](https://railway.com/deploy/rift-1) - Rift Its a OSS AI Chat for teams
- [ClingySOCKs](https://railway.com/deploy/clingysocks) - Deploy your own AI agents with relational memory engine instantly
- [Clawless](https://railway.com/deploy/clawless) - Clawless: Production-ready AI agents with durable state and structured UI.
- [RailClaw](https://railway.com/deploy/railclaw-1) - One-click OpenClaw deployment — persistent, containerized, just add keys.
- [vast-bold](https://railway.com/deploy/vast-bold) - OpenClaw-only instance for Lumina warm pool provisioning
- [pure-keen](https://railway.com/deploy/pure-keen) - DiaCheckAI: Smart early diabetes detection. Check, prevent, and manage now.
- [autofigure-edit](https://railway.com/deploy/autofigure-edit) - AutoFigure-edit turns paper methods into editable SVG figures
- [AutoDetailing SaaS Starter (AI + Cloudinary + MongoDB)](https://railway.com/deploy/autodetailing-saas-starter-ai-cloudinary) - Auto detailing CRM & SaaS starter with AI and image uploads
- [holyclaude](https://railway.com/deploy/holyclaude) - HolyClaude is a browser-capable Claude Code environment
- [AgentLogs](https://railway.com/deploy/agentlogs) - Collaboration in the age of agentic engineering.
- [copaw](https://railway.com/deploy/copaw) - Your Personal AI Assistant; supports multiple chat apps
- [Paperclip](https://railway.com/deploy/paperclip-1) - One-click persistent Paperclip on Railway with built-in setup flow.
- [aurogen](https://railway.com/deploy/aurogen) - Aurogen is a modular multi-agent platform inspired by OpenClaw
- [openocta](https://railway.com/deploy/openocta) - OpenOcta is an AI hub that combines gateway, agent runtime, channels, cron
- [torchcode](https://railway.com/deploy/torchcode) - A self-hosted, JupyterLab-based PyTorch interview practice platform
- [AEX Run](https://railway.com/deploy/aex-run) - Self-hosted AI-powered ERP.
- [ravishing-motivation](https://railway.com/deploy/ravishing-motivation) - Proxy server for Matchwiser AI widget
- [skillhub](https://railway.com/deploy/skillhub) - SkillHub is an enterprise-ready, self-hosted agent skill registry
- [aicomicbuilder](https://railway.com/deploy/aicomicbuilder) - AI Comic Builder is an open-source, AI-powered comic videos pipeline
- [Spacebot](https://railway.com/deploy/loyal-nurturing) - Deploy and Host Spacebot with Railway
- [openmaic](https://railway.com/deploy/openmaic) - OpenMAIC is an open-source AI multi-agent interactive classroom platform
- [Next.js + AI SDK](https://railway.com/deploy/nextjs-ai-sdk) - Next.js AI Chat Template using Vercel AI SDK
- [RepoIntel](https://railway.com/deploy/repointel-1) - Turn any GitHub repo into a searchable    intelligence layer.
- [EchoDeck](https://railway.com/deploy/echodeck) - Generate a mp4 from powerpoint with TTS
- [LocalAI](https://railway.com/deploy/localai-1) - Zero Config | One Click | Fully Persisted | Secured UI
- [cloudsecurity-af](https://railway.com/deploy/cloudsecurity-af) - AI Powered Cloud Security Analysis
- [Dialoqbase - AI Chatbot Builder | Open Source Chatbase Alternative on Railway](https://railway.com/deploy/self-host-dialoqbase) - Self-host DIaloqbase. Chat with your docs - PDF, URL, Git, YT
- [Arkenos](https://railway.com/deploy/arkenos) - Composable orchestration layer for conversational AI infrastructure.
- [openclaw with native onboard](https://railway.com/deploy/openclaw-with-native-onboard) - openclaw for Railway with a minimal reverse proxy wrapper.
- [OpenFang](https://railway.com/deploy/openfang) - A template for OpenFang: The Agent Operating System.
- [Detecting AI Pickling](https://railway.com/deploy/detecting-ai-pickl-1) - Reproducible environment testing static scanner on AI pickle-based files
- [claude-flux](https://railway.com/deploy/claude-flux) - Your personal AI assistant on Telegram, powered by Claude, runs on Railway.
- [Outbound Tools (Open Source Alternative To Lemlist, Instantly or Smartlead)](https://railway.com/deploy/outbound-tools) - Outbound email tools to send, tag, and manage via lightweight IMAP/SMTP
- [Inbox](https://railway.com/deploy/inbox) - Dead-simple todo manager for AI agents. SQLite + MCP, nothing else.
- [Slurm on Railway](https://railway.com/deploy/slurm-on-railway) - Play with a real Slurm scheduler on Railway
- [OpenClaw - Pre-configured and pre-hardened (For Non-Technical Users)](https://railway.com/deploy/openclaw-railway) - Security-hardened always-on OpenClaw AI personal assistant
- [Spacebot](https://railway.com/deploy/spacebot-1) - Openclaw alternative for teams, communities, and multi-user environments.
- [Spacebot (Slim)](https://railway.com/deploy/spacebot-slim) - An AI agent for teams, communities, and multi-user environments.
- [Teavision](https://railway.com/deploy/earnest-courage) - A tool for tea tasters to classify tea regions based on tea liquor color
- [OpenClaw AI + n8n workflows + Tailscale security](https://railway.com/deploy/openclaw-ai-n8n-workflows-tailscale-secu) - OpenClaw AI Agent Stack — n8n , Tailscale Mesh, Modal & 500+ Integrations
- [Background Agent](https://railway.com/deploy/background-agent) - Deploy and host coding agents (opencode/claude code/codex) on Railway
- [consultralph](https://railway.com/deploy/passionate-abundance) - ConsultRalph - AI autonomous agent for consultants
- [OpenClaw (Clawdbot) - Railway Template](https://railway.com/deploy/openclaw-clawdbot-railway-template) - Deploy and Host OpenClaw (Clawdbot) - Railway Template with Railway
- [Dexter with Web UI](https://railway.com/deploy/dexter-with-web-ui) - Dexter : autonomous financial research agent with a secure web dashboard.
- [MongoDB Community Server Replica Set](https://railway.com/deploy/mongodb-community-server-replica-set) - Deploy a production-ready, high-availability MongoDB cluster
- [OpenClaw](https://railway.com/deploy/openclaw-2) - Production-ready OpenClaw template, fully maintained
- [CrawlChat 2026.02.05](https://railway.com/deploy/crawlchat-20260205) - Power up your documentation with AI - CrawlChat.app
- [claw-control](https://railway.com/deploy/claw-control) - Kanban for AI Agents - Coordinate your AI team with style
- [MadBull](https://railway.com/deploy/madbull) - Your own MadBull AI assistant with web dashboard and setup wizard.
- [secure-moltbot-railway-template](https://railway.com/deploy/secure-moltbot-railway-template) - Moltbot setup with Cloudflare Tunnel secured
- [Open Lovable](https://railway.com/deploy/open-lovable) - Deploy and host Open Lovable on Railway with ComputeSDK
- [Tabby AI Code Assistant](https://railway.com/deploy/tabby-ai-code-assistant) - TabbyML Self-hosted AI code completion server using OpenAI. tabby
- [LLM Stack](https://railway.com/deploy/llm-stack) - LiteLLM with Redis for Production
- [TrendRadar — Automated News Monitoring & Daily Intelligence](https://railway.com/deploy/trendradar-automated-news-monitoring-and) - trendradar, automated news crawling, summaries, RSS tracking, web reports.
- [LLMOps](https://railway.com/deploy/llmops) - Ship LLM features confidently with production-grade infrastructure.
- [AnythingLLM (Just Updated)](https://railway.com/deploy/anythingllm-latest) - Private document chat and RAG workspace, data kept on a volume
- [ChromaDB (Just Updated)](https://railway.com/deploy/chromadb-latest) - Vector database for embeddings, pinned image, data kept on a volume
- [Magic Resume](https://railway.com/deploy/magic-resume) - AI-powered resume builder with real-time preview, auto-save, and PDF export
- [Function Gemma](https://railway.com/deploy/function-gemma) - Lightweight function calling model based on Gemma 3
- [cloude-agent](https://railway.com/deploy/cloude-agent) - Deploy the Claude Agent SDK to the cloud. Invoke via API or webhooks
- [FastRTC](https://railway.com/deploy/fastrtc) - Deploy and Host FastRTC with Railway
- [vanna ai](https://railway.com/deploy/vanna-ai-template) - Deploy and Host vanna ai with Railway
- [lightrag](https://railway.com/deploy/lightrag) - One-click Graph RAG • Knowledge Graph + Vector Search • Web UI included
- [TezosX-mcp](https://railway.com/deploy/tezosx-mcp) - An MCP server enabling AI agents to interact with the Tezos blockchain.
- [withoutBG](https://railway.com/deploy/withoutbg) - AI Image Background Removal
- [OpenMonetize](https://railway.com/deploy/openmonetize) - Open-source AI usage metering and billing platform.
- [stellar-optimism](https://railway.com/deploy/stellar-optimism) - i made an api for this project
- [Graphiti + Neo4j](https://railway.com/deploy/graphiti-neo4j) - Build Real-Time Knowledge Graphs for AI Agents
- [Llana (Open-Source LLM Management & AI Workflow Platform)](https://railway.com/deploy/llana) - Llana [Sep ’26] (Run, Chain & Manage AI Models Easily) Self Host
- [AI Powered Analytics](https://railway.com/deploy/ai-powered-analytics) - Analyses keywords on google and ai platforms.
- [Open Agent Builder](https://railway.com/deploy/open-agent-builder) - Visual AI agent workflow builder powered by Firecrawl
- [Apollo MCP Server](https://railway.com/deploy/apollo-mcp-server) - Deploy-ready Apollo MCP server for Apollo MCP projects
- [ZenML (Open-Source MLOps Framework & Pipeline Orchestration Tool)](https://railway.com/deploy/zenml) - ZenML [Sep ’26] (ML Pipelines, MLflow & Airflow alternative) Self Host
- [QueryWeaver](https://railway.com/deploy/queryweaver) - Translate natural language into SQL with a graph-powered tool.
- [open-webui-mcpo-n8n](https://railway.com/deploy/open-webui-mcpo-n8n) - Deploy and Host Open Webui with n8n using MCP for full stack AI Automation
- [Context7](https://railway.com/deploy/context7) - Up-to-date code documentation for LLMs and AI code editors
- [Embedding Gemma](https://railway.com/deploy/embedding-gemma) - Generate high-quality text embeddings with Google's best model
- [gpt-vLLM](https://railway.com/deploy/gpt-vllm) - Deploy and host your own LLM with VLLM
- [fireplexity](https://railway.com/deploy/fireplexity) - Fast AI search engine with citations and streaming responses.
- [Deepseek-R1-8b](https://railway.com/deploy/deepseek-r1-8b) - Deploy and host Deepseek-R1-8b on railway.
- [gpt-vLLM (old)](https://railway.com/deploy/something-oss) - Deploy and Host LLM with VLLM
- [InnerLink: nextjs better-auth postgres](https://railway.com/deploy/innerlink-nextjs-better-auth-postgres) - AI powered Therapy with AI-Human Therapist
- [mem0-chatbot](https://railway.com/deploy/mem0-chatbot) - Personalised AI chatbot that retains information with the help of Mem0
- [vygil-ai](https://railway.com/deploy/vygil-ai) - Autonomous Activity Tracking & Anomaly Detection
- [nextchat](https://railway.com/deploy/nextchat) - ✨ Light and Fast AI Assistant template for Railway
- [LangServe Neo4j Symantec Layer ](https://railway.com/deploy/langserve-neo4j-symantec-layer) -  A template to host a LangServe server backed by a Neo4j Symantec Layer.
- [Railway AI Docs Search (Cohere + pgvector)](https://railway.com/deploy/railway-ai-docs-search-cohere-pgvector) - Semantic search engine for your documents, using Cohere embeddings
- [rundown](https://railway.com/deploy/rundown-1) - AI-powered RSS reader that helps you read less and learn more
- [FlowiseAI with Workers](https://railway.com/deploy/flowiseai-with-workers) - Flowise AI with Workers - persisted volume, PostGIS & private networking.
- [Masumi Compliant Service API Official](https://railway.com/deploy/masumi-compliant-service-api-official) - Deploy Masumi-compliant AI agent wrapper
- [Kortix SUNA AI](https://railway.com/deploy/kortix-suna-ai) - Deploy and Host Kortix SUNA AI with Railway
- [cf5 chat](https://railway.com/deploy/cf5-chat) - Chat with multiple models in a sleek interface.
- [Open-webui with Artifacts](https://railway.com/deploy/open-webui-with-arti) - the famous AI Chat UI but easier to work with codes -- similar to v0.dev
- [AI Agentic Automation Stack (n8n + Flowise)](https://railway.com/deploy/ai-agentic-automation-stack-n8n-flowise) - All-in-One AI Automation Stack: workflows, agents, vector DB, web scraping
- [zooming-contentment](https://railway.com/deploy/-98HdG) - AI Coach that helps to improve pronunciation
- [N8N MCP Server](https://railway.com/deploy/se2WHK) - Give your AI Agents access to your n8n. Create workflows and more!
- [Milvus](https://railway.com/deploy/c7nLmV) - Authenticated Milvus v2.6 REST API with etcd and MinIO.
- [adk-starter](https://railway.com/deploy/eVpydG) - set up to use googles adk
- [zealous-enthusiasm](https://railway.com/deploy/1UXt2K) - Banco de dados do Trade unico e exclusivo
- [FLUJO](https://railway.com/deploy/7XqZC-) - Orchestrates AI models, MCP servers, and workflows via a single interface.
- [gitingest-js](https://railway.com/deploy/cVfQZ1) - Chat with any git repo, with image and pdf processing
- [Scira MCP AI](https://railway.com/deploy/KPUOlm) -  An open-source AI chatbot app powered by Model Context Protocol (MCP)
- [Ragpi Recaptcha Gateway](https://railway.com/deploy/scntIP) - A Web Widget for the Ragpi AI Assistant
- [Ragpi Discord Integration](https://railway.com/deploy/cgiAdr) - A Discord bot for the Ragpi AI Assistant
- [Bedrock Access Gateway](https://railway.com/deploy/H-Ax3g) - OpenAI-Compatible RESTful APIs for Amazon Bedrock
- [Open WebUI with MCP](https://railway.com/deploy/xB7OU2) - Open WebUI with MCP-to-OpenAPI proxy server
- [supallm](https://railway.com/deploy/48SIfr) - Build AI Agents with no code and control them using our realtime SDK.
- [Dify (Mail: Resend / ETL:Unstructured / VDB:Weaviate)](https://railway.com/deploy/t05D2s) - Dify works with Resend + Unstructured + Weaviate
- [nextjs AI-Chatbot](https://railway.com/deploy/NckAGE) - A full-featured, hackable Next.js AI chatbot built by Vercel
- [Extract Chat - chat with firecrawl](https://railway.com/deploy/UzxNSc) - Scrape anything with a simple chat  [powered by Firecrawl & OpenAI]
- [sqlchat](https://railway.com/deploy/KsHjEY) - Chat-based SQL Client and Editor for the next decade
- [firecrawl:v1.13.0](https://railway.com/deploy/-iHken) - Scrape, crawl and extract with a single API.
- [superglue](https://railway.com/deploy/wJg5-W) - Self-healing open source data connector.
- [Flowboard360 Setup Template - Flowise](https://railway.com/deploy/2XlruV) - Self hosted dashboard for your Flowise chatbots
- [Ragpi Slack Intgration](https://railway.com/deploy/aaWjR4) - A Slack bot for the Ragpi AI Assistant
- [Ragpi](https://railway.com/deploy/7ihedX) - 🤖 An AI assistant answering questions from your documentation
- [FlowiseAI + PostgreSQL](https://railway.com/deploy/oBCbmw) - FlowiseAI + PostgreSQL Database
- [N8N + Worker + Webhook + External DB](https://railway.com/deploy/Gos_2q) - n8n with Separate Worker & Webhook using External DB
- [karakeep:v0.30.0](https://railway.com/deploy/lGB4F_) - Quickly save links, notes, and images
- [Open DeepResearch [by firecrawl]](https://railway.com/deploy/s9kj-n) - AI agent that deeply researches the web, using firecrawl and Reasoning LLM
- [ElizaOS](https://railway.com/deploy/aW47_j) - Deploy your own AI agent that can interact across clients using ElizaOS.
- [spark-stack-template](https://railway.com/deploy/61t3xx) - Spark Stack is an tool for building web apps with prompts.
- [OpenAI Voice Assistant](https://railway.com/deploy/pCNq15) - Extensible voice assistant template built with the Realtime API and FastAPI
- [UndetectableAI-Coder](https://railway.com/deploy/Akvll8) - A bolt template made by Faisal for UndetectableAI
- [FlowiseAI + MySQL](https://railway.com/deploy/XcRehu) - FlowiseAI + MySQL Database
- [Verba (RAG Chatbot)](https://railway.com/deploy/QFa7Oq) - Retrieval Augmented Generation (RAG) chatbot powered by Weaviate
- [Bolt.DIY](https://railway.com/deploy/YaqTB9) - Build, run, and deploy full-stack web apps using any LLM of your choice!
- [bolt.diy](https://railway.com/deploy/C--NRV) - AI-Powered Full-Stack Web Development in the Browser
- [N8N (AI Starter Kit)](https://railway.com/deploy/n8n-ai-starter-kit) - A powerful workflow automation tool for technical people
- [MindsDB](https://railway.com/deploy/otiqwz) - MindsDB is the platform for building AI from enterprise data
- [FlowiseAI 🇧🇷](https://railway.com/deploy/DC2bT-) - Template simples Flowise AI. Defina usuário, senha e faça o deploy.
- [Morphic AI](https://railway.com/deploy/I7-4De) - Perplexity-like answer engine powered with AI,
- [Interactive_Chatbot](https://railway.com/deploy/HqQk1G) - This is AI Chatbot you need to experience
- [openui](https://railway.com/deploy/NzJO6n) - Deploy OpenUI: AI-powered UI generation with GitHub OAuth and OpenAI API.
- [FastAPIChat](https://railway.com/deploy/_-qAbG) - A FastAPI App for benchmarking LLM models from various vendors.
- [langflow 1.0-alpha w/ user auth](https://railway.com/deploy/Z1CH2j) - Authenticated langflow sessions only.
- [Librechat with RAG](https://railway.com/deploy/cnhjS_) - Librechat.ai with RAG enabled
- [Langflow Pre-releases](https://railway.com/deploy/UsJ1uB) - Simple Langflow 1.0 Preview deployment using PostgreSQL as the database.
- [ChatGPT-to-API](https://railway.com/deploy/Yxd_qk) - Create a fake API using ChatGPT's website
- [Langtrace](https://railway.com/deploy/8dNq1c) - Monitor, evaluate & improve your LLM apps!
- [LanguageTool](https://railway.com/deploy/KdtxnQ) -  More than a Grammar Checker
- [Unburn Toys](https://railway.com/deploy/K1iD7P) - Everyday AI Tools: Unburn Toys
- [coze2openai](https://railway.com/deploy/yM5tQL) - Use Coze on your favorite OpenAI client.
- [OpenAI Assistants LInk](https://railway.com/deploy/-M8zbi) - Link your OpenAI Assistants to a database + Evaluate Assistant responses
- [Chainlit](https://railway.com/deploy/atS4DW) - A production-ready Conversational AI applications framework in Python. 
- [astra-assistants-chat-ui](https://railway.com/deploy/v2UVYh) - DataStax Open Source GenAI Starter
- [ollama](https://railway.com/deploy/tXERGO) - ollama container with a demo/testing interface
- [Magika](https://railway.com/deploy/nnnGdy) - An AI powered file type detection tool
- [Rivet + Chatbot-UI](https://railway.com/deploy/XjMVyQ) - Hosts Rivet with an Open AI compatible API + Chatbot-UI as a frontend
- [MLflow Tracking](https://railway.com/deploy/R1yjwG) - The MLflow Tracking is an API and UI for logging ML experiment.
- [SwiftOpenAIProxy](https://railway.com/deploy/ocPcV2) - Server-side Swift OpenAI reverse proxy server
- [Label Studio (Single Service)](https://railway.com/deploy/X_ZJld) - Deploys Label Studio via a single docker image
- [Promptmodel](https://railway.com/deploy/uzaE0r) - Host, engineer and manage prompt & model configurations.
- [ollama-webui](https://railway.com/deploy/k4_cpH) - Web UI for Ollama, frontend for LLMs
- [LangServe - Pirate Speak Template](https://railway.com/deploy/pW9tXP) - Basic Langserve server with the pirate speak template
- [zep-memory](https://railway.com/deploy/ugib3R) - Conversational LLM Application Framework
- [flowise-nomocodes](https://railway.com/deploy/fUNRIK) - A template for deploying FlowiseAI with persisted volume.
- [Sentence Transformers Model](https://railway.com/deploy/oFLL6X) - FastAPI server to serve sentence-transformer embeddings
- [Standalone `pgvector` Service](https://railway.com/deploy/uYYYQy) - A PostgreSQL instance with `vector` extension enabled.
- [Pinecone](https://railway.com/deploy/Rg70kF) - Sample Pinecone apps for document summarization and question-answering.
- [Mage AI](https://railway.com/deploy/Cr5cge) - 🧙 A modern replacement for Airflow.
- [LangChain Apps](https://railway.com/deploy/oE8sWo) - Sample LangChain apps for search, text, document, URL, and news summaries.

Open this page in a browser: https://railway.com/deploy/category/ai