OpenAI-compatible router that puts free LLM models from 13 providers behind one endpoint with automatic failover
Open-source LLMOps platform for engineering teams
Stage
Not given
MVP
Founded
Not given
Not given
Based in
FR
Not given
Pricing model
Freemium
Open source
Pricing
Free: open core under BSL 1.1, no account or card required. Pro: 12 € monthly (or 9.08 € per month billed annually at 109 €/yr), or 199 € for Lifetime license (one payment, no renewals, limited to first 200). 14-day free trial for Pro requires a card.
Free during beta. Self-hosted is free under Apache 2.0 license. Enterprise options available on request.
Key features
Automatic failover across 13 free LLM providers
OpenAI-compatible API endpoint
Model capability testing for tools, JSON and vision
Hardcoded zero paid model fallback
Local router with keys staying on your machine
Model fusion and answer synthesis in Pro
Custom virtual model profiles in Pro
Native prompt compression to reduce tokens
Automatic model cooling and recovery probing
Support for Cursor, Continue, Open WebUI, LangChain and Claude Code
Prompt versioning with named aliases and diffs
Server-side prompt templating with conditionals and loops
OpenAI-compatible AI gateway with automatic model fallbacks
Virtual keys, budgets and per-key rate limits
LLM tracing through the gateway or over OpenTelemetry
Versioned tool catalog for function calling
Offline experiments and online evaluation rules
Prompt optimization with candidate rewrites
Team-wide audit trail of every change
MCP server for Claude Code, Claude Desktop, Cursor
What makes it different
Structured inability to spend money on paid models, enforced in code not configuration
No middleman: keys stay local, prompts go directly to providers
Benchmarked quality parity with gpt-5 for classification, extraction and reasoning
Millisecond failover when a model rate-limits or fails
Change prompts without a deploy by moving production alias
Ficelle
AcruxCore