HN Showcase
Back to browse
GitHub Repository

Multi-AI MCP Server - Connect ChatGPT, Claude, Gemini & Perplexity to your coding tools without any API

1,091 starsJavaScript

Proxima – local open-source multi-model MCP server (no API keys)

by Zen4bit·Feb 17, 2026·2 points·0 comments
Visit ProjectView on HN

AI Analysis

●●SolidNiche GemShip It
The Take

Turns your browser-signed sessions with ChatGPT, Claude, Gemini and Perplexity into one localhost endpoint (localhost:3210) with Python/JS SDKs, a smart router and 45+ MCP tools like file analysis and math search. It's a clever pragmatic play — useful tooling (REST shim, tool palette, retry logic) — but the no-API-key angle is inherently brittle and raises reliability/observability and privacy tradeoffs, so expect engineering work to keep it stable.

Target Audience

Developers and AI power users who want to prototype multi-model workflows, maintain local privacy, or unify access to multiple LLM providers

Post Description

I built Proxima, an open-source MCP server that runs locally and lets you orchestrate multiple AI providers in one workflow. I used it to run a multi-model “dev team” experiment (planning → coding → review → fixes). Looking for feedback on architecture, MCP tool design, and reliability/observability.

If you want, I can tailor the title to be more HN-friendly based on whether you’re submitting as link post (GitHub) or text post (write-up).

Similar Projects

Infrastructure●●Solid

Bifrost: Fastest enterprise AI gateway

Bifrost combines an OpenAI-compatible front door with adaptive load balancing, semantic caching, automatic failover, cluster mode and a built-in web UI — you can spin it up with npx or Docker in seconds. The performance claims (sub-100µs overhead at 5k RPS, '50x faster than LiteLLM') and multi-provider routing are the project's selling points; I want to see independent benchmarks and deeper docs on guardrails/provider quirks before trusting it for critical workloads.

WizardrySolve My ProblemSlick
aanthonymax
104mo ago