⚡
Performance-Benchmark
Wie schnell ist das Modell?
Roher Durchsatz auf echter Hardware – wie viele Token ein Modell pro Sekunde erzeugt, wie schnell es den Prompt verarbeitet und wie kurz die Zeit bis zum ersten Token ausfällt. Je höher, desto besser.
⚡ Generation (tok/s)🔄 Prefill⏱️ TTFT👥 Concurrency
| # | Modell / Hersteller | Messwerte | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | gemma-4-E2B-it5BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 903,30 tok/s TG Prefill 6.960 · TTFT 11.382 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | gemma-4-E2B-it5BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 860,88 tok/s TG Prefill 6.641 · TTFT 11.413 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 488,18 tok/s TG Prefill 4.229 · TTFT 20.915 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 4 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 484,77 tok/s TG Prefill 4.124 · TTFT 21.088 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 5 | gemma-4-E2B-it5BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 478,55 tok/s TG Prefill 6.243 · TTFT 3.601 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 6 | gemma-4-E2B-it5BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 461,54 tok/s TG Prefill 5.838 · TTFT 3.795 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 7 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 455,06 tok/s TG Prefill 3.954 · TTFT 21.269 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 8 | Nemotron-3-Nano-4B4BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 367,75 tok/s TG Prefill 3.397 · TTFT 29.121 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 9 | Nemotron-3-Nano-4B4BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 354,29 tok/s TG Prefill 3.544 · TTFT 30.045 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 10 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 253,67 tok/s TG Prefill 1.397 · TTFT 48.511 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 11 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 251,39 tok/s TG Prefill 3.708 · TTFT 6.599 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 12 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 250,66 tok/s TG Prefill 1.396 · TTFT 48.986 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 13 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 247,76 tok/s TG Prefill 1.399 · TTFT 49.349 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 14 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 246,68 tok/s TG Prefill 3.165 · TTFT 7.126 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 15 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 245,62 tok/s TG Prefill 3.740 · TTFT 6.690 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 16 | Nemotron-3-Nano-4B4BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 210,25 tok/s TG Prefill 3.483 · TTFT 8.798 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 17 | Nemotron-3-Nano-4B4BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 202,30 tok/s TG Prefill 3.149 · TTFT 9.380 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 18 | gpt-oss-20b20BOpenAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 202,12 tok/s TG Prefill 4.412 · TTFT 4.498 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliMXFP4 | Details → | |
| 19 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 182,87 tok/s TG Prefill 7.295 · TTFT 5.014 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 20 | MiniMax-M2.7230BMiniMax PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 174,20 tok/s TG Prefill 5.042 · TTFT 47.620 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawAWQ | Details → | |
| 21 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 169,89 tok/s TG Prefill 7.470 · TTFT 5.007 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 22 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 167,11 tok/s TG Prefill 2.814 · TTFT 6.937 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 23 | gpt-oss-20b20BOpenAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 148,88 tok/s TG Prefill 4.312 · TTFT 2.296 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliMXFP4 | Details → | |
| 24 | Nemotron-Cascade-2-30B-A3B30BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 148,79 tok/s TG Prefill 3.362 · TTFT 6.588 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 25 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 146,91 tok/s TG Prefill 4.736 · TTFT 2.900 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 26 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 142,76 tok/s TG Prefill 1.244 · TTFT 17.133 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 27 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 140,35 tok/s TG Prefill 1.237 · TTFT 17.315 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 28 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 139,82 tok/s TG Prefill 1.244 · TTFT 17.279 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 29 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 139,24 tok/s TG Prefill 5.091 · TTFT 2.946 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 30 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 135,61 tok/s TG Prefill 2.590 · TTFT 3.731 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 31 | Nemotron-Cascade-2-30B-A3B30BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 123,54 tok/s TG Prefill 2.897 · TTFT 3.743 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 32 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 108,98 tok/s TG Prefill 1.446 · TTFT 66.475 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 33 | MiniMax-M2.7230BMiniMax PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 106,12 tok/s TG Prefill 5.634 · TTFT 12.854 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawAWQ | Details → | |
| 34 | MiniMax-M2.5230BMiniMax PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 103,49 tok/s TG Prefill 7.016 · TTFT 5.294 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawAWQ | Details → | |
| 35 | Nemotron-Cascade-2-30B-A3B30BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 101,36 tok/s TG Prefill 3.237 · TTFT 6.850 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 36 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 101,18 tok/s TG Prefill 6.144 · TTFT 314 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 37 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 100,59 tok/s TG Prefill 6.798 · TTFT 285 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 38 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 92,96 tok/s TG Prefill 4.802 · TTFT 404 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 39 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 90,05 tok/s TG Prefill 1.600 · TTFT 43.239 ms | 10× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 40 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 86,23 tok/s TG Prefill 2.963 · TTFT 824 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 41 | Nemotron-Cascade-2-30B-A3B30BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 84,61 tok/s TG Prefill 2.849 · TTFT 3.806 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 42 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 84,18 tok/s TG Prefill 2.987 · TTFT 816 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 43 | gpt-oss-20b20BOpenAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 82,16 tok/s TG Prefill 3.589 · TTFT 550 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliMXFP4 | Details → | |
| 44 | MiniMax-M2.5230BMiniMax PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 77,74 tok/s TG Prefill 5.460 · TTFT 2.641 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawAWQ | Details → | |
| 45 | Nemotron-Cascade-2-30B-A3B30BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 76,95 tok/s TG Prefill 2.356 · TTFT 929 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 46 | Nemotron-3-Nano-4B4BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 76,87 tok/s TG Prefill 2.223 · TTFT 1.006 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 47 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 74,19 tok/s TG Prefill 4.923 · TTFT 393 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 48 | Nemotron-3-Nano-4B4BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 72,70 tok/s TG Prefill 1.798 · TTFT 1.470 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 49 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 68,40 tok/s TG Prefill 2.292 · TTFT 842 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 50 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 63,50 tok/s TG Prefill 2.576 · TTFT 769 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 51 | Nemotron-Cascade-2-30B-A3B30BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 62,04 tok/s TG Prefill 2.341 · TTFT 935 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 52 | gemma-4-E4B-it8BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 60,50 tok/s TG Prefill 1.862 · TTFT 1.599 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 53 | Nemotron-3-Nano-30B-A3B30BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 58,25 tok/s TG Prefill 5.466 · TTFT 395 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 54 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 49,10 tok/s TG Prefill 1.157 · TTFT 10.523 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 55 | glm-4.7-flashZ.ai (Zhipu) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 48,92 tok/s TG Prefill 4.055 · TTFT 536 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawAWQ | Details → | |
| 56 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 48,48 tok/s TG Prefill 1.201 · TTFT 2.031 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 57 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 48,44 tok/s TG Prefill 1.189 · TTFT 2.051 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 58 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 47,20 tok/s TG Prefill 1.176 · TTFT 2.071 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 59 | Gemma-4-26B-A4B26BGoogle PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 47,06 tok/s TG Prefill 4.378 · TTFT 451 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 60 | Qwen3.5-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 46,91 tok/s TG Prefill 3.713 · TTFT 537 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawFP8 | Details → | |
| 61 | Qwen3.6-35B-A3B35BQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 46,79 tok/s TG Prefill 3.315 · TTFT 589 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawFP8 | Details → | |
| 62 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 46,50 tok/s TG Prefill 1.156 · TTFT 10.550 ms | 5× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 63 | Qwen3-Coder-NextQwen (Alibaba) PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 38,54 tok/s TG Prefill 2.124 · TTFT 1.146 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawFP8 | Details → | |
| 64 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 33,12 tok/s TG Prefill 1.171 · TTFT 2.084 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 65 | MiniMax-M2.5230BMiniMax PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,23 tok/s TG Prefill 2.184 · TTFT 1.050 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawAWQ | Details → | |
| 66 | MiniMax-M2.7230BMiniMax PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 31,52 tok/s TG Prefill 2.328 · TTFT 991 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawAWQ | Details → | |
| 67 | Ling-3.0-flash127.5BinclusionAI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 28,27 tok/s TG Prefill 1.151 · TTFT 2.118 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 68 | Nemotron-3-Super-120B-A12B120BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,39 tok/s TG Prefill 1.495 · TTFT 1.449 ms | 1× | NVIDIA GB10 (DGX Spark)10 x ARM implementer 0x41, part 0xd87 + 10 x ARM implementer 0x41, part 0xd85 · 120 GB RAM | vLLMgodclawNVFP4 | Details → |