Provider pricing

Hyperbolic AI API pricing

Hyperbolic is a serverless inference provider focused on full-precision (BF16) serving of open-weight flagships — Llama 4, DeepSeek V4, Qwen3, FLUX. Pricing is among the lowest in the market, and the platform also exposes a GPU marketplace for fine-tuning and custom workloads.

Founded2023
HeadquartersSan Francisco, CA
Tracked modelsDirectory

Best for

Quality + cost-sensitive teams that want full-precision open-weight inference.

Strengths

  • Full BF16 precision — no quantization quality loss
  • Among the lowest per-token rates on Llama 4 and DeepSeek
  • GPU marketplace for renting H100/H200 capacity
  • OpenAI-compatible API with streaming