Fireworks AI
74.0Trending upFireworks AI|Rank #85
The fastest inference platform for deploying and fine-tuning open-source AI models at production scale with pay-per-token pricing.
80.0
Performance
25.0
Popularity
80.0
Value
55.5
Trust
Overview
Fireworks AI is an inference platform founded in 2022 by former Meta and Google AI veterans. It provides blazing-fast serverless inference for hundreds of open-source models across text, image, audio, and multimodal. Processing over 10 trillion tokens daily for 10,000+ customers, it raised $250M at a $4B valuation in October 2025. Fireworks offers serverless pricing, dedicated GPU deployments, and reinforcement fine-tuning with sub-second latency.
Pricing Plans
Free Credits
Free
- Free credits to start
Serverless
Custom
- From $0.20/M tokens
- Auto-scaling
- 40% batch discount
Enterprise
Custom
- Volume discounts
- SLA
- Dedicated GPU
Features
Sub-second inference
100s of open-source models
Serverless & dedicated GPU
Fine-tuning with RL
OpenAI-compatible API
Quick Info
- Category
- Open Source
- Status
- active
- Monthly Visits
- 200.0K
- API Available
- Yes
- Open Source
- No
