Fireworks vs Together AI
13 vs 34 models, 10 shared
Shared models
| Model | Fireworks $/1M out | Fireworks tok/s | Together AI $/1M out | Together AI tok/s |
|---|---|---|---|---|
| MiniMax-M2.7 | 166 tok/s | 155 tok/s | ||
| MiniMax-M3 | 95 tok/s | 19 tok/s | ||
| DeepSeek-V4-Pro | 47 tok/s | 57 tok/s | ||
| Kimi-K2.6 | 92 tok/s | 176 tok/s | ||
| Kimi-K2.7-Code | 120 tok/s | 54 tok/s | ||
| NVIDIA-Nemotron-3-Ultra-550B-A55B-NVFP4 | 138 tok/s | 149 tok/s | ||
| gpt-oss-120b | 182 tok/s | 107 tok/s | ||
| gpt-oss-20b | 83 tok/s | 143 tok/s | ||
| GLM-5.1 | 45 tok/s | 54 tok/s | ||
| GLM-5.2 | 51 tok/s | 111 tok/s |