Loading...
Loading...
Models ranked by Output Speed (tok/s), based on independent benchmark evaluations.
Each row shows the model's benchmark score alongside its pricing and output speed, so you can evaluate quality-to-cost tradeoffs at a glance.
Top 20 models ranked by output speed (tok/s)
| Rank | Model | Output Speed (tok/s) |
|---|---|---|
| 🥇 | Inception | 777 tok/s |
| 🥈 | Alibaba | 422 tok/s |
| 🥉 | Alibaba | 364 tok/s |
| 4 | Alibaba | 357 tok/s |
| 5 | 353 tok/s | |
| 6 | 347 tok/s | |
| 7 | Alibaba | 342 tok/s |
| 8 | IBM | 342 tok/s |
| 9 | IBM | 339 tok/s |
| 10 | NVIDIA | 319 tok/s |
| 11 | 315 tok/s | |
| 12 | Amazon | 309 tok/s |
| 13 | Mistral | 295 tok/s |
| 14 | 288 tok/s | |
| 15 | OpenAI | 285 tok/s |
| 16 | NVIDIA | 284 tok/s |
| 17 | 259 tok/s | |
| 18 | OpenAI | 254 tok/s |
| 19 | OpenAI | 244 tok/s |
| 20 | Alibaba | 244 tok/s |