Overview

Unlock an unprecedented scale of free AI inference with FreeLLMAPI. This project aggregates the generous free tiers of 34 different LLM providers, offering access to 635 distinct model endpoints through a single, unified /v1 API. It seamlessly integrates with any OpenAI-compatible client, providing a powerful solution for personal experimentation without the costs associated with proprietary APIs. Smart routing, automatic failover, and encrypted key management ensure a robust and secure experience, allowing you to leverage billions of tokens per month at no charge.

Key Features

• Massive Free Token Pool: Access an aggregate of 7.4 billion tokens per month by combining the free tiers of numerous LLM providers.
• Unified API Access: Interact with 635 free model endpoints, plus any custom OpenAI-compatible endpoint, through a single /v1 API, dramatically simplifying integration.
• Intelligent Routing & Failover: A sophisticated router automatically selects the best available model for each request, providing seamless failover to the next provider when one is rate-limited or unavailable.
• Broad Compatibility: Works with all OpenAI-compatible clients, including popular CLIs, coding agents like Claude Code and Codex CLI, and development tools. Supports Anthropic Messages API and native Gemini/Ollama surfaces.
• Dynamic Model Catalog: The router auto-updates its model catalog from a signed feed, ensuring you always have access to the latest free models, quota changes, and compatibility fixes without manual intervention. Free installs receive a monthly snapshot, with premium routers getting immediate updates.
• Enhanced Security: Your provider API keys are stored encrypted at rest using AES-256-GCM and only decrypted in memory per request. Your applications interact via a single, unified freellmapi-… bearer token.
• Comprehensive Features: Includes support for image, video, and audio generation, tool calling, structured outputs, prompt compression, sticky sessions, and a rich admin dashboard with analytics.
• Local-First & Customizable: Deploy locally via Docker or native installers for Windows, macOS, and Linux. Offers self-hosting capabilities and a desktop menu-bar app for seamless local integration.

Typical Use Cases

• Personal AI Experimentation: Explore a vast array of LLMs and their capabilities for creative projects, learning, and prototyping without incurring API costs.
• Development & Testing: Integrate LLM functionality into applications for testing purposes or for internal tools where cost is a significant factor.
• Prototyping & MVP Development: Quickly build and iterate on AI-powered features for new applications, leveraging extensive free inference capacity.
• Educational Purposes: Students and educators can access powerful AI models for research, coursework, and hands-on learning about LLM technology.
• AI Agent Integration: Power coding assistants, chatbots, and other AI agents with a broad spectrum of LLM models through a unified interface.