Welcome Fireworks.ai on the Hub 🎆

Hugging Face Blog · 1 year ago

Fireworks.ai is now available as a supported inference provider on Hugging Face Hub, allowing users to run serverless inference on models including DeepSeek-R1, DeepSeek-V3, Mistral-Small, and Llama-3.2-90B. Users can access Fireworks inference through the Hub's website UI, Python and JavaScript SDKs, or HTTP calls, with billing passed through directly at standard Fireworks API rates and no markup from Hugging Face. PRO subscribers receive $2 in monthly inference credits that work across all providers on the platform.