#1717th of 48 in Gateways and observability
axonhub
LLM gateway that translates between OpenAI, Anthropic and Gemini APIs
- Stars
- 5.5k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Nov 2026
- Language
- Go
3,135 on 21 Apr 20265,452 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
AxonHub is a Go gateway that accepts requests in OpenAI, Anthropic or Gemini format and translates them for the configured provider channel, so a client SDK can call another vendor's model without code changes. It adds request tracing, RBAC with usage quotas, per-request cost tracking and load balancing with failover. Embedding, rerank and image generation endpoints are also supported.
Who it is for: Teams routing multiple LLM providers through one API with quotas and tracing
Strengths
- Cross-format translation: call Claude with the OpenAI SDK or GPT with the Anthropic SDK
- Per-request cost breakdown for input, output and cached tokens, plus thread-level tracing
- Covers text, image generation, embedding and rerank APIs across about a dozen providers
- Public demo instance with a demo account for trying the admin UI
Weaknesses
- Realtime API is listed as Todo; AWS Bedrock and Google Cloud channels are still in testing
- Mixed licensing: Apache-2.0 plus LGPL-3.0 for the llm directory
- Deployment is delegated to an agent skill; the README gives no manual install steps
- Main README is mostly Chinese, with sponsor blocks taking much of the page
What it needs
- no GPU
- Docker + Compose
- Compose runs PostgreSQL
- Models: OpenAI, Anthropic, Gemini, DeepSeek, Zhipu
- port 8090
Quick answers
Is axonhub open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run axonhub with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Does axonhub need a GPU?
No, it runs without one.
Also in Gateways and observability
See all 48| Rank | Project | Score |
|---|---|---|
| 1 | OmniRouteLLM gateway routing one OpenAI-compatible endpoint across 372 providers | 96 out of 100 |
| 2 | headroomCompresses tool outputs, logs and history before they reach the LLM | 92 out of 100 |
| 3 | new-apiGateway that unifies LLM providers behind OpenAI, Anthropic and Gemini APIs | 91 out of 100 |
| #4 | MLflowTracing, evaluation and prompt registry for LLM apps, plus ML tracking | 90 out of 100 |
| #5 | LiteLLMOpenAI-format gateway and Python SDK for calling 100+ LLM providers | 87 out of 100 |