Agent runtime infrastructure Multimodal GPU Infrastructure Inference Optimization Enterprise Adoption Model serving Model Routing Coding Agents Benchmarks Reinforcement Learning Kimi

Fireworks AI

The fastest way to build with generative AI

Visit site

Generative AI inference and deployment platform for accessing, fine-tuning, and serving open models via APIs and managed infrastructure.

Recent stories

1 linked story

releaseSECONDARY2026-07-08

Cognition launches SWE-1.7 in Devin at 1,000 tok/s

Cognition says SWE-1.7 was trained with RL on a Kimi K2.7 base and now runs in Devin at 1,000 tok/s. It reports 42.3% on FrontierCode at $1.97 per task and released revised grading rules.