Fireworks partners with the world's leading generative AI researchers to serve the best models, at the fastest speeds. Production AI platform designed to collaborate and serve the models, at high speeds.The company's platform aims to run, fine-tune, and share large language models (LLMs) to solve product problems, enabling developers and businesses to scale at a faster rate through rapid product iteration building as well as minimizing cost to serve.

Fireworks.ai offers generative AI platform as a service. We optimize for rapid product iteration building on top of gen AI as well as minimizing cost to serve.

Models curated and optimized by Fireworks

Mixtral MoE 8x7B Instruct : Mistral MoE 8x7B Instruct v0.1 model with Sparse Mixture of Experts.

FireFunction V1 : Fireworks' open-source function calling model.

Llama 2 70B Chat : A fine-tuned version of Llama 2 70B, optimized for dialogue applications using Reinforcement Learning from Human Feedback (RLHF), and perform comparably to ChatGPT according to human evaluations.

Mistral 7B Instruct : The Mistral-7B-Instruct-v0.1 Large Language Model (LLM) is a instruct fine-tuned version of the Mistral-7B-v0.1 generative text model using a variety of publicly available conversation datasets.