Fireworks Serverless
Call open models like Llama, DeepSeek, Qwen and gpt-oss through a fast, OpenAI-compatible API, paying per token.
Overview
Fireworks Serverless runs a large catalogue of open models on Fireworks' own inference engine. You call them through an OpenAI-compatible API and pay per million tokens from prepaid credits, with no servers or GPUs to manage. Most open models follow simple size-based prices per million tokens: under 4B parameters US$0.10, 4B to 16B US$0.20 and over 16B US$0.90; mixture-of-experts models up to 56B cost US$0.50 and 56B to 176B US$1.20. Newer models have their own prices, for example gpt-oss-120b at US$0.15 input and US$0.60 output. Batch jobs cost half the serverless price. Fine-tuned models (LoRA or full fine-tuning) are served at the same price as their base model. Prices change often, so check Fireworks' pricing page for current rates.
Specifications
| Hosting type | Serverless API |
|---|---|
| Model families | Llama, Mistral, DeepSeek, Qwen, Gemma, gpt-oss |
| LLM models | Llama 3.3 70B Instruct, Mistral NeMo, DeepSeek-R1, DeepSeek-V3.1, Qwen2.5 72B Instruct, Qwen2.5-Coder 32B Instruct, Qwen3 32B, Qwen3 30B-A3B, Qwen3 235B-A22B, Gemma 2 9B, Gemma 3 4B, Gemma 3 12B, Gemma 3 27B, gpt-oss-120b, gpt-oss-20b |
| Pricing range | US$0.05 – US$22.5 |
| Pricing model | Pay per token |
| Free tier available | No |
| Data center locations | |
| Hardware / GPU options | Fully managed. Models run on Fireworks' GPU fleet; you don't choose hardware. |
| VRAM & memory | |
| Maximum model size | 671B |
| OpenAI-compatible API | Yes |
News about Fireworks AI
All news →Fireworks AI: serverless and on-demand hosting now in our directory
We've added Fireworks AI, the fast open-model inference platform founded by Meta's former PyTorch leads, with its serverless and on-demand hosting plans and 17 supported models.
