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

Fireworks AI

Fast, scalable infrastructure for generative AI.

Visit site

Fireworks AI is a hosted generative-AI platform/API for deploying, serving, and fine-tuning open and proprietary models in production, emphasizing fast inference, lower serving cost, and workflows for building AI applications and agents.

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.