Fireworks AI

Fireworks AI is a platform for ultra-fast inference of generative models, letting teams run state-of-the-art open-source LLMs and image models with low latency. It provides built-in fine-tuning and deployment (including their RFT offering) so customers can customize models — often at no additional cost — and achieve performance that can rival frontier models. The product is aimed at developers and companies seeking cost-effective, high-performance model hosting and easy model customization and scaling.