A free, open-source, OpenAI-compatible gateway that pools the free tiers of 24 cataloged LLM providers.
freellmpool is a free, open-source Python tool that pools the free tiers of 24 LLM
providers — Groq, Cerebras, NVIDIA, Google Gemini, OpenRouter, Cloudflare, Mistral, Cohere and more —
behind one OpenAI-compatible endpoint. It runs as a command-line tool, a Python library, a
local proxy, and an MCP server, with automatic failover
when a provider is rate-limited and per-day quota tracking across tiers. Several providers (Pollinations,
OVHcloud, and Kilo Gateway) need no API key, and LLM7 works without one, so pip install freellmpool can answer without signup when a keyless provider is up.
The proxy speaks the OpenAI API and includes an experimental Anthropic-compatible path, so existing apps
and coding agents such as Codex, Claude Code, and aider can run on pooled free models.
pip install freellmpool
freellmpool ask "Explain the CAP theorem in one sentence." # keyless when routes are availableGitHub · PyPI · Changelog · MIT license
spread routing. To test the main-only additions explicitly:
python -m pip install --force-reinstall 'git+https://github.com/0xzr/freellmpool.git@main'opencode-freellmpool and
opencode-freellmpool-tui are CI-tested but were not published on npm as of 2026-07-19;
use their repository-local installation paths.| Tool | What it is | Install | Keyless start | CLI / library / proxy / MCP |
|---|---|---|---|---|
| freellmpool | Pools many providers' free tiers | pip install | Yes (keyless routes) | All four |
| OpenRouter | Hosted paid aggregator (some free models) | API key | No (account + key) | API only |
| LiteLLM | Multi-provider SDK/proxy (bring your own keys) | pip install | No | Library + proxy |
| Self-hosted free-API servers | A server you deploy | Docker + config | No (deploy first) | Server only |
freellmpool's niche: the keyless, pip-installable client for squeezing the hosted free tiers — not a server you deploy and not a paid aggregator.
freellmpool ask "...", pipe stdin, pin a provider/model.OPENAI_BASE_URL at freellmpool proxy for most text, embedding, and transcription calls.freellmpool code claude (also codex, aider, cline, continue, cursor, Hermes, and OpenCode) routes them to pooled free models — see the step-by-step guide. The Hermes profile is currently main-only.main adds public /livez and /readyz probes, authenticated /v1/providers, and /v1/models?ready=true for locally ready targets.freellmpool/spread for quota-aware agent loops and add the repository-local dashboard/tools plugins while npm publication is pending.from freellmpool import Pool, AsyncPool.freellmpool mcp lets Claude Desktop/Code/Cursor offload subtasks to free models.freellmpool benchmark + latency-aware routing.freellmpool capacity status and freellmpool providers health show which free tiers are healthy, near quota, or missing a key — see the capacity guide.Per-provider free tiers: Groq · Gemini · Cerebras · NVIDIA · Mistral
Yes — freellmpool is a free, open-source, MIT-licensed gateway that exposes one OpenAI-compatible
endpoint backed by the free tiers of 24 cataloged providers. Install it with pip install freellmpool
and point any OpenAI client at the local proxy.
freellmpool pools them: each request goes to a provider you have access to, and if that one is
rate-limited or down it automatically fails over to the next. It tracks per-day usage so load spreads
evenly across the free tiers, and an optional fast routing mode prefers the
lowest-latency provider.
Yes. The proxy implements the OpenAI API and an experimental Anthropic-compatible path, so you set
OPENAI_BASE_URL=http://localhost:8080/v1 or ANTHROPIC_BASE_URL=http://localhost:8080 and run Codex,
Claude Code, aider, Cline, Continue, Cursor, or OpenCode on pooled free models.
Run freellmpool code claude for the exact setup. (The Claude Code path is experimental:
text and tool use, no vision yet.)
No key is required to start: Pollinations, OVHcloud, and Kilo Gateway need no API key, and LLM7 is key-optional, so a fresh install can answer without signup when a keyless provider is up. Add free keys for the other providers to unlock more models and higher limits.
24 cataloged providers: Pollinations, OVHcloud, LLM7, Kilo Gateway, OpenCode Zen (disabled by default), Groq, Cerebras, NVIDIA NIM, OpenRouter, Google Gemini, GitHub Models, Cloudflare Workers AI, Mistral, Cohere, SambaNova, Z.ai/Zhipu, Ollama Cloud, LongCat, Hugging Face (router), Aion Labs, ModelScope API Inference, Morph, Vercel AI Gateway, and SiliconFlow — 247 enabled chat routes, 405 cataloged chat models, plus free embeddings and audio transcription (Whisper).
Run freellmpool capacity status: it labels each provider healthy,
low_quota, exhausted, invalid_key, or missing from
your local catalog, environment, and per-day usage, and flags when you're below a target number of
healthy providers. freellmpool providers health sends one tiny request to each configured
provider and reports latency or failure, and freellmpool keys add walks you through
configuring more. See the capacity guide.
OpenRouter is a hosted aggregator you pay per token (with some rate-limited free models) and access
through its API. freellmpool is a local, open-source tool you pip install that pools the
providers' own free tiers directly, can start keyless when routes are available, and ships a CLI, a Python library, a
proxy, and an MCP server.
Yes — it is MIT-licensed and free to use. Source is on GitHub; the package is on PyPI.
Free-tier models are smaller than frontier models — good for drafting, classification, triage, and everyday coding, not GPT-class reasoning on the hardest problems. Limits reset at UTC midnight and capacity varies through the day. The proxy is meant for local/single-user use.
freellmpool is not a privacy layer: prompts go to the selected upstream provider. See the FAQ for provider routing, ToS, and privacy notes.