GoModel
A self-hosted AI gateway written in Go that puts OpenAI, Anthropic, Gemini, Ollama and dozens of other providers behind one compatible API.
GoModel at a glance
What is GoModel?
GoModel sits between your applications and model providers. It accepts requests in OpenAI format at /v1 and in Anthropic format at /v1/messages, so the official SDKs keep working once their base URL points at the gateway. Behind that endpoint it adds exact and semantic response caching, per-request cost estimates, hard budgets per user, team or key, rate limits, failover with circuit breakers, model allowlists and guardrails. A dashboard shows spending, live requests and a playground, and the gateway also exports Prometheus metrics and OpenTelemetry traces. It can aggregate MCP servers behind one endpoint too.
Good fit if
- You pay for several model providers and want spending caps and usage reports in one place.
- Your apps already use the OpenAI or Anthropic SDKs and you want to add caching or backup providers without code changes.
- You prefer a single Go binary or container over a Python service.
Look elsewhere if
- You want a chat interface for people rather than an API layer: Open WebUI or LibreChat.
- You need a gateway with a longer track record and a Python SDK: LiteLLM.
- You want to serve open models on your own GPUs: vLLM.
Before you adopt it
The gateway is MIT-licensed and no paid code was found in the repository. ENTERPILOT also sells GoModel Pro, a separate commercial build that adds prompt compression, intelligent routing and OIDC single sign-on for the dashboard. The project is young: the repository dates from December 2025 and the latest release is v0.2.1. Ten people committed in the last 90 days, but one author wrote most of that code. Docker images, builds for Windows, Linux and macOS, and a Helm chart are available; the full Compose stack uses Redis, PostgreSQL and MongoDB.
Drafted with AI from the project's README and repository data, then checked against that data on 31 October 2026. Report a correction
How to run GoModel yourself
One-click install
Not listed in the app catalogs of Umbrel, YunoHost, TrueNAS, Unraid, CasaOS, Runtipi, Proxmox VE Helper-Scripts and Coolify.
Ways to install
- Docker files
- Dockerfile or compose file in the repository
- Docker image
- enterpilot/gomodel: 125k pulls · x86-64, ARMv7, ARM64 · 24 MB download
- Compose file
- docker-compose.yaml: 6 services (redis, postgres, mongodb)
- Setup effort
- 3 of 5, judged from the compose file
- Kubernetes
- Helm chart in the repository
- Packages
- AUR versions on Repology
- Desktop and mobile
- Builds for macOS, Linux, Windows, including ARM in GitHub releases
- Interface languages
- 4 languages: Chinese, English, German and Polish
Install instructions: gomodel.enterpilot.io or the README on GitHub. Docker images and packages are checked weekly.
GoModel health: is it still maintained?
83 of 100, actively developed
Read from GitHub on 12 November 2026. How the score works
GoModel security
Security advisories are the vulnerabilities the maintainers disclosed on GitHub, with the versions that fix them; many advisories can mean careful disclosure as much as weak code.
Latest advisories
Advisories read from GitHub on 12 November 2026.
GoModel license: is it open source?
GoModel uses MIT, an open-source license approved by the OSI.
- December 2025First published on GitHub.
- TodayMIT.
GoModel for developers
Written in
- Go 84%
- JavaScript 10%
- Svelte 4%
- Shell 1%
- CSS <1%
- Other <1%
Getting involved
Language shares are GitHub’s count of the code in the repository.
Alternatives to GoModel
- LiteLLMThe fastest, litest AI Gateway. Rust core with Python SDK. Call 100+ LLM APIs in OpenAI (or native) format… 83 health · MIT · 63.2k stars
- OllamaGet up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models. 89 health · MIT · 187k stars
- Open WebUIUser-friendly AI Interface (Supports Ollama, OpenAI API, ...) 72 health · Open WebUI License · 158k stars
- LibreChatEnhanced ChatGPT Clone: Features Agents, MCP, Skills, DeepSeek, Anthropic, AWS, OpenAI, Responses API,… 85 health · MIT · 47k stars
- Local Deep Research~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google,… 84 health · MIT · 9.6k stars
Questions about GoModel
Is GoModel open source?
Yes. GoModel is published under MIT, an OSI-approved open-source license.
Is GoModel still maintained?
Its health score is 83 of 100 (actively developed). It had 551 commits in the last 90 days, the latest on 12 November 2026, and its latest release came out on 8 November 2026.
How do I run GoModel?
You can run it with the Docker image enterpilot/gomodel, on Kubernetes with the Helm chart in the repository and as an app for macOS, Linux, Windows, including ARM. The README on GitHub has the exact steps.
What is GoModel written in?
GoModel is written mainly in Go (84% of the code), with JavaScript (10%), by GitHub’s count of the repository.
What languages is GoModel available in?
The repository has interface translations for 4 languages: Chinese, English, German and Polish.
What are the alternatives to GoModel?
Close alternatives in the AI chat & local LLMs category include LiteLLM, Ollama, Open WebUI.