freellmpool
Active·★ 129·MIT·Updated 2026-10-16
freellmpool aggregates the free tiers of 19 LLM providers into a single OpenAI-compatible endpoint, enabling access to diverse models without API keys for some. It handles automatic failover between providers, tracks daily usage, and offers various interfaces including a CLI, Python library, and local proxy.
#LLM Proxy#Free Tier#OpenAI API#API Gateway#Failover#Usage Tracking#Python Library#CLI Tool
01
Features
01OpenAI-compatible and experimental Anthropic-compatible API endpoint for LLMs and other services
02Automatic failover and load balancing across multiple free LLM providers on rate limits, timeouts, or errors
03Per-day usage tracking and intelligent quota management for free tiers to maximize availability
04Keyless operation for several providers, allowing immediate use without API key setup
05Intelligent routing based on prompt difficulty, latency, or usage (e.g., quality, fast, fair modes)
02
Compatibility
Python
Runtime
Verified via docs
OpenAI API
API Compatibility
Verified via docs
Anthropic API
API Compatibility
Verified via docs
CLI
Interface
Verified via docs
03
Quick start
1
$ pip install freellmpool
04
Use cases
↳Run coding agents (e.g., Claude Code, aider) against pooled free LLM models locally
↳Benchmark and monitor the health and capacity of various free LLM providers in real-time
↳Integrate free LLM capabilities into Python applications with a simple library API for tasks like summarization or embeddings
↳Serve as a local OpenAI-compatible proxy for existing tools and scripts, redirecting requests to free tiers
↳Queue and manage slow, quota-aware LLM jobs for background processing and reporting
05
Alternatives
harnessrouter★ 3.2k
HarnessRouter Community Edition: the self-hosted, Apache-2.0 edition of the unified interface for agent harnesses. Run Codex, Claude Code, Hermes, PI, DSH, and more through one API, with sessions, streaming, files, cancellation, and failure handling. Implements the Unified Harness Protocol (UHP), an open standard. Your keys, your infrastructure.
all2md★ 33
Convert PDF, Word, PowerPoint, HTML, email & 40+ formats to clean Markdown — and back. Built for LLMs, RAG & Python pipelines, with a built-in MCP server.
docker-claudebox★ 23
Claude Code in Docker. Drop-in OpenAI-compatible API, MCP server, Telegram bot, and CLI — five interfaces, one image. Persistent sessions, file ops, always-on skill injection, and a full dev toolchain (Go, Python, Node, K8s, Terraform, databases) or a minimal image with just the basics.
conduit★ 235
Local MCP gateway that cuts your agent's tool-token overhead ~90%. Every MCP server's tools collapse to 3 meta-tools the agent searches on demand. Measured: 97% less per request, same results. Keys in your OS keychain, no cloud.
llmtrim★ 255
Local proxy that compresses your LLM API requests so you pay less, with no change to the answers. Trims wasted tokens from prompts, history, tool output, and code before they're sent: -31% input / -74% output, measured live. Any provider, no extra model calls. Also an MCP server and embeddable library (Rust, Python, Ruby, Kotlin, Swift).
Comments
Log in to leave a comment
No comments yet. Be the first!