Provider pricing

Lambda AI API pricing

Lambda is best known for its on-demand and reserved GPU cloud, and its Inference API exposes a curated catalog of open-weight LLMs — Llama 4, Qwen3, DeepSeek V4, Hermes — at aggressive per-token rates. The same account also unlocks 1-Click Clusters and reserved H100/H200/B200 capacity for training and custom inference.

Founded2012
HeadquartersSan Jose, CA
Tracked modelsDirectory

Best for

AI teams that want managed inference plus owned GPU capacity from a single vendor.

Strengths

  • Per-token serverless inference plus on-demand and reserved GPUs
  • Tight integration with Lambda's GPU cloud for fine-tuning and training
  • Competitive pricing on Llama 4 and DeepSeek flagships
  • OpenAI-compatible API with streaming and tool use