--- title: "Deploy Typesense Elixir" description: "community Elixir client against Railway" category: "Analytics" url: https://railway.com/deploy/typesense-elixir --- # Deploy Typesense Elixir community Elixir client against Railway **[Deploy Typesense Elixir on Railway](https://railway.com/template/typesense-elixir)** Machine-readable deploy manifest (JSON, validated by TemplateCI): https://railway.com/deploy/typesense-elixir/manifest.json - **Creator:** onepush - **Category:** Analytics ## Template content ### typesense-railway - **Source:** Shinyduo/typesense-railway - **Public domain:** Yes ## Documentation # Deploy and Host self hosted Typesense Elixir (Open-Source Instant Search) on Railway Phoenix search feels instant until the SaaS invoice arrives. Typesense Elixir points your Elixir/OTP app at self-hosted Typesense on Railway — `typesense/typesense:30.2` on port 8108 with `/data`, `TYPESENSE_API_KEY`, and `--enable-cors`. ## About Hosting Typesense Elixir open-source software on Railway (self hosted Typesense template) Typesense is a single Go binary you drop onto Railway, point a Phoenix app at, and get faceted, typo-tolerant search in an afternoon. The Elixir pattern drives the Typesense HTTP API using Req or HTTPoison, then feeds results into a LiveView. One container, one volume, one API key. If you call Typesense server-side, CORS doesn’t matter; browser InstantSearch requires `--enable-cors`. ## Why Deploy Typesense Elixir, the Algolia alternative on Railway (Railway Free Trial) Algolia bills per search request and per record stored — a moderately popular site can cost hundreds monthly. Typesense charges only compute and storage. On Railway, a small instance with a 1–2 GB volume costs single-digit to low-teens USD. The $5 GitHub trial covers your first month, so you can benchmark for free. You own the index, typo tolerance, and data export; Algolia doesn’t allow self-hosting. Railway is a singular platform to deploy your infrastructure stack. Railway will host your infrastructure so you don't have to deal with configuration, while allowing you to vertically and horizontally scale it. By deploying Typesense Elixir on Railway, you are one step closer to supporting a complete full-stack application with minimal burden. Host your servers, databases, AI agents, and more on Railway. ### Railway vs Other Hosting Providers and VPS for Typesense Elixir self hosting Railway sits between VPS and PaaS: one-click volume, automatic HTTPS, private networking — no Terraform. Less control over VM specs, but managed Postgres and private network often beat raw CPU. ## Common Use Cases for hosted Typesense Elixir A Phoenix product catalog with search-as-you-type, indexing Postgres data via Oban and calling Typesense from LiveView with Req. Full-text search over user content — forum posts, docs, tickets — using `num_typos: 2`. Browser-side InstantSearch on a static site through Railway’s public HTTPS domain, with Phoenix only serving the page. ## Dependencies for Typesense Elixir Docker hosted on Railway You need the official Typesense container on Railway plus an Elixir HTTP client in `mix.exs`. Pin `typesense/typesense:30.2`, mount `/data`, set `TYPESENSE_API_KEY`, and start with `--data-dir /data --api-key=$TYPESENSE_API_KEY --enable-cors` on port 8108. ### Deployment Dependencies for Managed Typesense Service (Instant Search) Pin image `typesense/typesense:30.2`. Start command: `--data-dir /data --api-key=$TYPESENSE_API_KEY --enable-cors`. Attach volume at `/data`. Generate a long random `TYPESENSE_API_KEY` (e.g., `openssl rand -hex 32`) in Railway shared variables — never lose it; recovery means deleting the volume and reindexing. No external database needed. ### Implementation Details for Typesense Elixir (Using Typesense official docker image) Use Req with Finch for LiveView concurrency. Define a thin module like `MyApp.Search.TypesenseClient`. Base URL for server-side calls: `http://typesense:8108` on Railway private network. Wrap calls in `try/rescue` with a 2-second timeout. CORS only matters for browser InstantSearch; restrict origins via reverse proxy in production. ## How does Typesense Elixir compare against other Instant Search platforms ### Typesense Elixir vs Algolia (Algolia Alternative) Algolia has polished docs and dashboard, but closed SaaS with per-request pricing that punishes success. At 1M requests/month you pay $100+; Typesense on Railway stays flat around $10–15/month for a small node. You lose managed infra and A/B testing, but for a Phoenix team it’s a fair trade. ### Typesense Elixir vs Meilisearch (Meilisearch Alternative) Meilisearch is Rust, disk-based with memory mapping; Typesense keeps everything in RAM. For small datasets Meilisearch runs on less RAM, but Typesense has lower query latency under heavy writes. Meilisearch’s API keys are more granular; both are callable from Elixir via Req. For millions of docs on a 2 GB VPS, Meilisearch may be more forgiving; for speed on a few hundred thousand, Typesense wins. ### Typesense Elixir vs Typesense Cloud (Typesense Cloud Alternative) ### Typesense Elixir vs Solr (Solr Alternative) Solr still wins for exotic analyzers and huge enterprise corpora, but it is a heavier JVM ops surface. Typesense Elixir keeps Phoenix search on a single Go binary with typo tolerance and facets, without standing up a SolrCloud. ## How to use Typesense Elixir (the OSS Instant Search)? Run Typesense as a separate service (Docker image `typesense/typesense:30.2` on port 8108 with data in `/data` and `--enable-cors`). Set `TYPESENSE_API_KEY` in your Elixir app environment, then start the Phoenix server. Open the app in a browser and use the search box to query your Typesense collection. The Elixir app calls Typesense over HTTP and renders results via LiveView. ## How to self host Typesense Elixir on other VPS Services (Typesense Elixir self hosting guide) Self-hosting Typesense Elixir on any VPS requires running both the Typesense server and the Elixir application. The steps below work on DigitalOcean, Linode, Hetzner, or any provider with a public IP. ### Clone the Repository Clone the Typesense Elixir repository from GitHub to your VPS. Use `git clone` and navigate into the project directory. ### Install Dependencies Install Elixir, Erlang, and Node.js (for asset compilation) on the VPS. Run `mix deps.get` and `mix assets.deploy` to fetch and compile dependencies. ### Configure Environment Variables Set `TYPESENSE_API_KEY` to match the API key you configure for the Typesense server. Also set `TYPESENSE_HOST` and `TYPESENSE_PORT` (default 8108) if Typesense runs on a separate host or port. ### Start the Typesense Elixir Application Run `mix phx.server` in the project directory after starting the Typesense container. The app listens on port 4000 by default; use a reverse proxy like Caddy or Nginx for HTTPS. ## Official Pricing of Typesense Elixir (Typesense Elixir pricing) Typesense Elixir itself is free and open-source under GPL-3.0. You pay only for infrastructure: Typesense Cloud starts at about $21.60/month for a 0.5GB burst instance. Self-hosting on a VPS costs single-digit to low-teens USD per month on Railway, plus a one-time $5 GitHub trial credit for new users. ## Typesense Elixir cloud vs self hosted comparison (Pricing, features, costs, and more) Typesense Cloud handles updates, backups, and scaling for a fixed monthly fee. Self-hosting gives full control and lower cost at small scale but requires managing the Typesense container and VPS. Feature parity is identical because Typesense Cloud runs the same open-source server. ### Monthly cost of self hosting Typesense Elixir on Railway Railway charges per resource: a small app container plus a Typesense volume typically totals single-digit to low-teens USD per month. Add a few dollars for persistent storage if your dataset grows beyond the free tier. ### System Requirements for Hosting Typesense Elixir on a VPS A VPS with 1 vCPU, 1GB RAM, and 20GB disk runs both Typesense and the Elixir app for small datasets (under 100k records). Typesense alone needs at least 512MB RAM; the Elixir app adds 200-300MB. Increase RAM to 2GB for larger collections or higher query throughput. ## Frequently Asked Questions (FAQs) ### Should I use HTTPS or localhost for Typesense Elixir? Use `localhost` for local development. For production, always expose Typesense and the Elixir app over HTTPS because browsers block mixed content and API keys would be exposed in plain HTTP. ### What happens if I lose the Typesense API key? Set a new `TYPESENSE_API_KEY` value in your environment and restart the Typesense container. Your existing data remains intact because the key only controls access, not encryption. ### Why do I need --enable-cors for LiveView? The `--enable-cors` flag lets the browser-based frontend call Typesense directly from a different origin. LiveView over websockets does not need CORS, but the search API calls from client-side JavaScript do. ### How much RAM does Typesense Elixir need? Typesense requires at least 512MB RAM for the server process. The Elixir app adds roughly 200MB, so plan for 1GB total on the same VPS for small workloads. ### Should I pin Typesense to version 30.2? Yes, pin the Docker image to `typesense/typesense:30.2` for reproducible behavior. Newer minor versions may change defaults or API responses, breaking the Elixir integration without warning. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. Keep the Typesense data volume and API key; InstantSearch hits port 8108 so queries stay fast under load. ## Similar templates - [Typesense PHP](https://railway.com/deploy/typesense-php) — official PHP client against Railway - [Typesense vs Meilisearch](https://railway.com/deploy/typesense-vs-meilisearch) — self-hosted Typesense vs Meilisearch - [Betterlytics](https://railway.com/deploy/betterlytics) — Betterlytics is a cookieless analytics platform GDPR-compliant. Open this page in a browser: https://railway.com/deploy/typesense-elixir