Provider pricing

Chutes AI API pricing

Chutes runs an OpenAI-compatible inference endpoint backed by Bittensor's decentralized GPU subnet, serving DeepSeek, Qwen, Llama, GLM, and Kimi at some of the lowest published rates on the market. Pricing is metered per token with no minimums and a generous free tier for evaluation.

Founded2024
HeadquartersDistributed (Bittensor)
Tracked modelsDirectory

Best for

Cost-sensitive teams that need cheap, scalable access to open-weight models without infra ops.

Strengths

  • Frontier open-weight models (DeepSeek V3.2, Qwen3, Kimi K2) at sub-cent pricing
  • OpenAI-compatible API — drop-in replacement for OpenAI/Together SDKs
  • Decentralized Bittensor backend, no per-model rate limits
  • Generous free credits for new accounts