Fireworks AI
Inference platform for running and tuning open AI models
fireworks.aiRedwood City, United States
Fireworks AI is a Redwood City company founded in 2022 by former Meta engineers led by Lin Qiao, who worked on PyTorch. It runs an inference and fine-tuning platform for open models and has trained the Ember model.
- Founded
- 2022
- Type
- Private company
- Founders
- Lin Qiao, Dmytro Dzhulgakov, Benny Chen, Chenyu Zhao, Dmytro Ivchenko, James Reed and Pawel Garbacki
- AI lab
- Inference provider
- 20 hosted endpoints
| Model | Endpoint | Context | Input /M | Output /M | Cached input /M | Cache write /M | Uptime 1d |
|---|---|---|---|---|---|---|---|
 DeepSeek V4 Flash 0423 | fireworks | 1M | $0.105 | $0.21 | $0.021 | — | 65.07% |
| DeepSeek V4.1 Flash | fireworks | 1M | $0.30 | $1.20 | $0.006 | — | 99.29% |
| DeepSeek V4.1 Flash | fireworks/spot | 1M | $0.21 | $0.84 | $0.0042 | — | 96.00% |
| DeepSeek V4.1 Flash | fireworks/us | 1M | $0.45 | $1.80 | $0.009 | — | 99.51% |
| DeepSeek V4.1 Flash 1017 | fireworks | 1M | $0.30 | $1.20 | $0.006 | — | 99.57% |
| DeepSeek V4.1 Flash 1017 | fireworks/spot | 1M | $0.21 | $0.84 | $0.0042 | — | 90.53% |
| DeepSeek V4.1 Flash 1017 | fireworks/us | 1M | $0.45 | $1.80 | $0.009 | — | 100.00% |
| Ember-1 | fireworks | 1M | $3.00 | $15.00 | $0.30 | — | 100.00% |
 GLM 5.3 | fireworks | 1M | $1.40 | $4.40 | $0.26 | — | 98.54% |
| GLM 5.3 | fireworks/fast | 1M | $2.10 | $6.60 | $0.39 | — | 99.26% |
| GLM 5.3 | fireworks/spot | 1M | $0.98 | $3.08 | $0.182 | — | 92.97% |
| GLM 5.3 Flash | fireworks | 1M | $0.15 | $0.50 | $0.03 | — | 99.79% |
 Kimi K3 | fireworks | 1M | $3.00 | $15.00 | $0.30 | — | 99.73% |
| Kimi K3 | fireworks/fast | 1M | $4.50 | $22.50 | $0.45 | — | 66.30% |
| Kimi K3 | fireworks/us | 1M | $4.50 | $22.50 | $0.45 | — | 100.00% |
| Kimi K3.1 Pro | fireworks | 1M | $2.00 | $10.00 | $0.20 | — | 99.32% |
 MiniMax M3.1 Flash | fireworks | 1M | $0.15 | $0.60 | $0.015 | — | 92.74% |
 Nemotron 3.5 Lightning | fireworks | 262K | $0.05 | $0.20 | $0.01 | — | 97.33% |
 Qwen3 Reranker 8B | fireworks | 41K | $0.00 | $0.00 | — | — | 99.64% |