AxonHub: The Open-Source AI Gateway That Kills Vendor Lock-In
AxonHub is an open-source AI gateway that lets you call 100+ LLMs from any SDK with zero code changes. Built-in failover, load balancing, tracing, and cost control — self-hosted in minutes.
🚀 Want to deploy AxonHub yourself?
Docker configs, system requirements, and installation guides — all on one page.
View AxonHub Tool Page →Every AI team knows the pain: you build against the OpenAI SDK, and a month later you want Claude — or Gemini, or DeepSeek. Rewriting every call, swapping SDKs, and re-testing integrations eats days of engineering time. AxonHub is an open-source AI gateway that makes provider switching a configuration change instead of a refactoring project.
What Is AxonHub?
AxonHub is an all-in-one AI development platform written in Go. It sits between your application and every LLM provider — OpenAI, Anthropic, Gemini, Zhipu, Kimi, DeepSeek, Doubao, Fireworks, OpenRouter and more (14+ providers, 100+ models). Your app keeps using the SDK it already knows; AxonHub transparently translates the request to any provider's native API. Zero code changes.
Why Teams Choose AxonHub
| Problem | AxonHub Solution |
|---|---|
| 🔒 Vendor lock-in | Switch GPT-4 → Claude → Gemini in one config change |
| 🧩 Integration complexity | One API format for 14+ providers |
| 📊 Observability gap | End-to-end request tracing out of the box |
| 💸 Cost control | Per-request cost breakdown + budgets |
Core Features
- Any SDK → Any Model: Use the OpenAI SDK to call Claude, or the Anthropic SDK to call GPT.
- Smart Load Balancing: Automatic failover in under 100ms, always routing to the healthiest channel.
- Full Request Tracing: Thread-aware timelines to debug any request.
- Enterprise RBAC: Fine-grained permissions, usage quotas, and data isolation.
- Real-time Cost Tracking: Input, output, and cache tokens tracked per request.
- Multi-modal: Text generation, image generation, embeddings, and reranking.
Open Source and Transparent
AxonHub is Apache-2.0 licensed (with the llm/ directory under LGPL-3.0) and has earned 4,852 GitHub stars. The Docker image looplj/axonhub has been pulled over 192,000 times — it's battle-tested in production. You can even try the live demo at axonhub.onrender.com before installing anything.
Get Started
Deployment takes minutes with the official Docker image. System requirements are modest: 2 vCPU and 4GB RAM minimum (4 vCPU / 8GB recommended). Check the tool page for the ready-to-use docker-compose configuration.
🚀 Ready to break free from vendor lock-in?
Docker configs, system requirements, and installation guides — all on one page.
View AxonHub Tool Page →