The free-tier landscape moves fast. Here are strong models available free right now, pulled straight from the live catalog.
- Kimi K2.6 (CF) — free on Cloudflare Workers AI, up to 262K context.
- Qwen3-Coder 480B (Ollama) — free on Ollama Cloud, up to 262K context.
- Qwen3 Coder 480B (free) — free on OpenRouter, up to 1.0M context.
- Kimi K2.7 Code (CF) — free on Cloudflare Workers AI, up to 262K context.
- Gemini 3.5 Flash — free on Google AI Studio, up to 1.0M context.
- DeepSeek V4 Pro (NV) — free on NVIDIA NIM, up to 131K context.
- MiniMax M2.7 (NV) — free on NVIDIA NIM, up to 197K context.
- Mistral Large 3 675B (NV) — free on NVIDIA NIM, up to 131K context.
- Kimi K2.6 (NV) — free on NVIDIA NIM, up to 131K context.
- Qwen3-Coder Next (Ollama) — free on Ollama Cloud, up to 262K context.
This list regenerates from the live FreeLLMAPI catalog as providers add and retire models. Want them the day they land instead of on the free 30-day delay? Premium keeps your router live.
Use every one of these through one key. FreeLLMAPI is an open-source, self-hosted router that puts all these free tiers behind a single OpenAI-compatible endpoint and fails over when one is rate-limited. Browse the catalog or go live.