lianluo-esign

lianluo-esign / ferrogate

Public

FerroGate is an open-source AI gateway and reverse proxy written in Rust. It is designed to route, secure, monitor, and control traffic to LLM providers such as OpenAI, Anthropic, Google Gemini, Azure OpenAI, and OpenAI-compatible APIs.

lianluo-esign.github.ioferrogate ai ai-gateway aigateway infrastructure new-api
11
0
100% credibility
Found May 06, 2026 at 11 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

FerroGate is a self-hosted gateway that routes AI chat requests to multiple providers with authentication, usage tracking, policies, and monitoring.

How It Works

1
🔍 Discover FerroGate

You find FerroGate, a helpful tool to manage and secure your AI chats from different services.

2
📥 Get it ready

Download the simple setup files to your computer.

3
🔗 Link your AI friends

Tell it which AI services like chat helpers you want to use by filling a short list.

4
🔑 Make private passes

Create special secure passes so only your team can use the AI chats.

5
▶️ Turn it on

Click start, and your secure AI hub comes alive on your machine.

6
💬 Chat away

Send messages to your AI through the hub, just like normal but safer.

📈 Watch it grow

Check the easy dashboard to see chats, usage, and everything running smoothly.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 11 to 11 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ferrogate?

FerroGate is a self-hostable Rust-based AI gateway and reverse proxy, designed as a control point for LLM traffic to providers like OpenAI, Anthropic, Google Gemini, Azure OpenAI, and compatible APIs. It handles routing with fallbacks, virtual API keys for secure access, policy enforcement like rate limits and token budgets, plus monitoring via structured logs, Prometheus metrics, and OTLP exports. Users get an OpenAI-compatible endpoint at /v1/chat/completions, admin dashboard, and automatic HTTPS via ACME, all configurable via familiar Caddyfile or TOML.

Why is it gaining traction?

Built on Cloudflare's Pingora for high-performance proxying, it stands out with production-ready features like circuit breakers, graceful reloads, and billing events—without the vendor lock-in of closed gateways. Devs appreciate the CLI for quick starts (cargo run -- run), config validation, and key hashing, plus Docker images for easy deployment. Multi-provider adapters and tenant-scoped policies make it a practical drop-in for controlling and monitoring LLM apis.

Who should use this?

AI platform engineers routing traffic across Anthropic, Gemini, and Azure need its fallback logic and observability. DevOps teams self-hosting LLM gateways will value the admin APIs for health checks and reloads. Startups tracking token usage for internal billing fit perfectly with its event sinks and aggregates.

Verdict

Early-stage with 11 stars and 1.0% credibility score, but MVP is validated end-to-end with strong docs, security audits, and tests—worth prototyping for Rust fans needing a ferrograd iron for LLM control. Skip for now if you require persistent storage or full CRUD admin.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.