Provider pricing

Featherless.ai AI API pricing

Featherless.ai flips the inference pricing model on its head: instead of per-token billing, you pay a flat monthly subscription for unlimited access to 4,000+ open-source models from Hugging Face — Llama, Qwen, DeepSeek, Mistral, and every popular community fine-tune. Cold-start times are typically under a second.

Founded2024
HeadquartersRemote / United States
Tracked modelsDirectory

Best for

Developers and hobbyists who want unlimited access to the open-source model zoo for a fixed monthly fee.

Strengths

  • Flat-rate plans from $10/month — predictable bill, unlimited tokens
  • 4,000+ Hugging Face models hot-swappable via OpenAI-compatible API
  • Sub-second cold starts on most 8B–70B class models
  • Ideal for roleplay, creative writing, and exploratory model evals