AxonHub is an all-in-one AI gateway that lets you call 100+ LLMs via any SDK with built-in failover, load balancing, cost tracking, and end-to-end tracing. It supports multiple deployment options and provides detailed config and usage guidance.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
AxonHub is an open-source AI gateway that enables using any SDK to call multiple language model providers (100+ LLMs) through a single API, with built-in failover, load balancing, cost control, and end-to-end tracing.
How it works
The project exposes a unified gateway API that translates SDK requests into provider-agnostic calls, enabling switching between providers without code changes. Features include intelligent load balancing, request tracing, and cost tracking. Supported providers include OpenAI, Anthropic, Zhipu, Moonshot, Gemini, and others via various APIs.
Getting started
30-Second Local Start
# 下载并解压（以 macOS ARM64 为例）
curl -sSL https://github.com/looplj/axonhub/releases/latest/download/axonhub_darwin_arm64.tar.gz | tar xz
cd axonhub_*
# 使用 SQLite 运行（默认）
./axonhub
# 打开 http://localhost:8090
# 首次运行：按照初始化向导设置系统（创建管理员账号，密码至少需要 6 位）
Zero-Code Migration Example
from openai import OpenAI
client = OpenAI(
base_url="http://localhost:8090/v1", # 指向 AxonHub
api_key="your-axonhub-api-key" # 使用 AxonHub API 密钥
)
response = client.chat.completions.create(
model="claude-3-5-sonnet", # 或 gpt-4、gemini-pro、deepseek-chat...
messages=[{"role": "user", "content": "Hello!"}]
)
Recent releases
Latest releases include v1.0.0-beta6 (2026-07-28) with UI/date display fixes, v1.0.0-beta5 (2026-07-11), v1.0.0-beta4 (2026-06-20), v1.0.0-beta3 (2026-06-11), and v1.0.0-beta2 (2026-06-02).
Traction
- Stars: 4868
- Forks: 629
- Open issues: 110
Deployment guides
- Docker Compose, Helm on Kubernetes, and Virtual Machine deployments are provided, including environment variable usage, configuration examples, and Helm chart values.
Usage and configuration
- Core config covers server, database (dialect), and logging options (example YAML and environment variables).
- Helm chart and deployment sections list key options such as axonhub.replicaCount, axonhub.dbPassword, and ingress/persistence toggles.
- Quick start and deployment steps include commands for cloning, setting env vars, and starting services.