Fireworks AI
Fireworks AI runs an inference platform for open models: serverless endpoints sold per token (with a priority tier, a fast serving mode and a US-pinned endpoint), on-demand deployments billed per GPU-second, and fine-tuning. Its price list names no region for the standard rows.
- Website: https://fireworks.ai
- Headquarters: Redwood City, United States
- Kind: Inference provider, Per-token API, GPU cloud
Fireworks AI sells 6 of the compared models per token and prices 5 kinds of GPU for hosting your own.
Models sold per token
| Model | Input, per 1M tokens | Output, per 1M tokens |
|---|---|---|
| DeepSeek V4.1 Flash | $0.30 | $1.20 |
| GLM 5.3 | $1.40 | $4.40 |
| GLM 5.3 Flash | $0.15 | $0.50 |
| Kimi K3 | $3.00 | $15.00 |
| MiniMax M3 | $0.30 | $1.20 |
| OpenAI gpt-oss 120B | $0.15 | $0.60 |
GPUs for hosting your own
| GPU | From, per GPU-hour | Cheapest instance | GPUs in it |
|---|---|---|---|
| H100 | $8.00 | h100.1x | 1 × 80 GB |
| H200 | $8.00 | h200.1x | 1 × 141 GB |
| B200 | $13.00 | b200.1x | 1 × 180 GB |
| B300 | $15.00 | b300.1x | 1 × 288 GB |
| GB300 | $20.00 | gb300.1x | 1 × 288 GB |
Where it runs
4 of its models are priced in a named place, in 1 region in 1 country; 2 have one global price and no stated place.
| Country | Regions | Per-token models | GPU kinds |
|---|---|---|---|
| United States | us | 4 | — |
No place stated for 2 models: one global price, and Fireworks AI does not say where a request is processed.
Where the prices come from
- Per-token prices: docs.fireworks.ai/serverless/pricing
- GPU servers: fireworks.ai/pricing
Sources
List prices collected 3 November 2026, refreshed every three hours. No negotiated rates or taxes.