Fireworks AI#

Documentation from Fireworks AI — fast inference and fine-tuning for open models.

Serverless and dedicated hosting for open-weight LLMs behind an OpenAI-compatible API, plus fine-tuning (LoRA, DPO, reinforcement fine-tuning). A peer of Together AI and DeepSeek for open-model inference.

Backward TRL - Transformers Reinforcement Learning Training Overview Forward