60benchmarks contributed
60Performance runs
0Harness runs
14Models tested
3GPUs used
2.264Best throughput · tok/s
Hardware used
Models tested
Top performance runs
| Model | Hardware | tok/s | TTFT | Date |
|---|---|---|---|---|
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B 10× | NVIDIA GeForce RTX 3090 Ti vLLM | 2.264 | 21.818 ms | Aug 12, 2026 |
| Qwen3.8-Flash-Next 10× | NVIDIA RTX A6000 | 1.172 | 45.396 ms | Oct 16, 2026 |
| NVIDIA-Nemotron-3.5-Lightning-30B-A3B 5× | NVIDIA GeForce RTX 3090 Ti vLLM | 1.135 | 9.812 ms | Aug 12, 2026 |
| Qwen3.6-35B-A3B 10× | NVIDIA GeForce RTX 3090 Ti | 831 | 14.928 ms | Jul 31, 2026 |
| Qwen3.6-35B-A3B 10× | NVIDIA GeForce RTX 3090 Ti | 830 | 14.891 ms | Jul 31, 2026 |
| Qwen3.6-35B-A3B 10× | NVIDIA GeForce RTX 3090 Ti | 828 | 14.940 ms | Aug 3, 2026 |
| Qwen3.8-Flash-Next 5× | NVIDIA RTX A6000 | 590 | 21.049 ms | Oct 16, 2026 |
| gpt-oss-20b 10× | NVIDIA GeForce RTX 3090 Ti vLLM | 572 | 1.915 ms | Aug 4, 2026 |
| Qwen3.6-35B-A3B 5× | NVIDIA GeForce RTX 3090 Ti | 467 | 5.143 ms | Aug 3, 2026 |
| gemma-4-E2B-it 10× | NVIDIA GeForce RTX 3090 Ti vLLM | 451 | 864 ms | Aug 3, 2026 |

