80+ free AI services for chat, image, video, voice & APIs (may sometimes include access to lead gen ai models for free)
Pool free tiers from 24 LLM providers behind one OpenAI-compatible API. 247 enabled routes, agent routing, failover, CLI, proxy, MCP, and OpenCode.
Zero-cost LLM routing across free-tier model APIs with Bifrost, certification scripts, and agent-ready fallback chains
Self-hosted OpenAI-compatible API proxy for Google Gemini — authenticated via OAuth 2.0 + PKCE, no paid API key required. Built with Bun + Hono + TypeScript.
Use Claude Code, Codex, OpenCode, Cline for free — one local gateway to 14+ free LLM providers. OpenAI & Anthropic compatible.
Daily-updated list of every free AI model available right now
How to Add Pollinations AI Models to OpenWebUI
Run Ollama on Google Colab with a public HTTPS endpoint via Cloudflare Tunnel or ngrok.
Control plane plugin for Hermes Agent — pool many free/cheap model providers + local AI CLIs, switch brains from Telegram in one command (even while the model is down), delegate work cap-aware with auto-fallback. Run your agent at ~$0 without hitting quota walls.
GLM-5.2 NVIDIA NIM Go 客户端 / OpenAI 兼容反向代理 — 自动化 hCaptcha 凭证池 + 流式推理，支持 Docker 部署。Go client & OpenAI-compatible proxy for NVIDIA Playground's GLM-5.2, with hCaptcha automation, captcha pool, streaming, and Docker.
GeminiHydra is an OpenAI-compatible API gateway that unlocks unlimited free-tier usage by aggregating multiple Google Gemini keys. Features automatic failover, rate-limit rotation, load balancing, and secure Cloudflare tunneling for remote access. includes a local dashboard for real-time monitoring.
Peak Intelligence
OpenAI-compatible proxy for ModelScope free LLMs. Auto-selects best model, failover on errors. Works with Cursor, Cline, Continue, Aider.
Use Claude Code 100% free with 100+ NVIDIA NIM models via LiteLLM proxy. No Anthropic subscription needed. Works on Windows & macOS.
Terminal coding agent that executes tools (real think>tool>result loop with confirm + undo, MCP tools in-loop) and rewrites its own system prompt from feedback with metrics + auto-rollback. Free by default via the FCM model router.
Make any LLM 10x smarter. Recursive DAG decomposition + temporal knowledge graph = free models that match expensive ones. 30/30 tasks passing at $0.0006 total cost.
🦆 No API key. No credit card. Turn the AI chat already open in your browser into a local OpenAI + Ollama endpoint for coding agents. Thread-aware — resends only what's new (16,531 → 523 chars per turn). Real SSE streaming, native function calling, multi-account failover. Self-hosted, MIT.
Fetches the highest-quality free OpenRouter models and automatically updates the Hermes Agent's default and fallback configurations whenever :free models expire.
Free, always-up LLM client & gateway for Python and TypeScript — one OpenAI-compatible API over free-tier OpenRouter, Google Gemini, NVIDIA NIM, Groq, Cerebras & Mistral with automatic failover, key rotation, streaming and live model discovery.