🏆
Ranglisten
LLM Leaderboard
Reale Messergebnisse aller getesteten Sprachmodelle auf echter, benannter Hardware – bewertet nach roher Geschwindigkeit (Performance in Token pro Sekunde, Prefill und Zeit bis zum ersten Token) und nach praktischer Aufgaben-Qualität in vollständigen Agenten- und Chat-Abläufen (Harness). Wähle unten einen Benchmark-Typ oder filtere nach Modell, Hersteller und Hardware, um genau zu sehen, was jeweils geprüft wird und wie die Ergebnisse zustande kommen.
⚡ Performance (tok/s)🤖 Harness-Qualität👥 Concurrency🖥️ echte Hardware
| # | Modell / Hersteller | Messwerte | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 411,74 tok/s TG Prefill 6.821 · TTFT 24.282 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 391,83 tok/s TG Prefill 5.175 · TTFT 25.101 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 342,61 tok/s TG Prefill 3.592 · TTFT 31.111 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 4 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 219,61 tok/s TG Prefill 6.206 · TTFT 7.079 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 5 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 217,53 tok/s TG Prefill 5.226 · TTFT 7.446 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 6 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 182,02 tok/s TG Prefill 3.427 · TTFT 9.736 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 7 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 143,98 tok/s TG Prefill 2.767 · TTFT 24.248 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 8 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 140,00 tok/s TG Prefill 2.638 · TTFT 27.354 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 9 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 138,38 tok/s TG Prefill 2.698 · TTFT 27.790 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 10 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 137,80 tok/s TG Prefill 3.136 · TTFT 23.970 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 11 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 115,88 tok/s TG Prefill 1.609 · TTFT 16.977 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 12 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 77,33 tok/s TG Prefill 2.366 · TTFT 6.171 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 13 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 77,23 tok/s TG Prefill 1.954 · TTFT 6.520 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 14 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 65,74 tok/s TG Prefill 1.560 · TTFT 8.148 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 15 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 64,59 tok/s TG Prefill 2.641 · TTFT 930 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 16 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 64,56 tok/s TG Prefill 3.452 · TTFT 709 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 17 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 57,37 tok/s TG Prefill 2.309 · TTFT 1.056 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 18 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 56,97 tok/s TG Prefill 2.087 · TTFT 6.098 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 19 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 45,16 tok/s TG Prefill 2.426 · TTFT 5.640 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 20 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 31,27 tok/s TG Prefill 1.242 · TTFT 1.972 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 21 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 31,24 tok/s TG Prefill 1.232 · TTFT 1.990 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 22 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 27,10 tok/s TG Prefill 1.215 · TTFT 2.015 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 23 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 23,25 tok/s TG Prefill 1.196 · TTFT 2.115 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 24 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 19,26 tok/s TG Prefill 1.308 · TTFT 1.876 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 25 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 19,26 tok/s TG Prefill 1.316 · TTFT 1.864 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 26 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,56 tok/s TG Prefill 50 · TTFT 190.018 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 27 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,48 tok/s TG Prefill 51 · TTFT 187.949 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 28 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,22 tok/s TG Prefill 50 · TTFT 189.758 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 29 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,13 tok/s TG Prefill 51 · TTFT 187.699 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 30 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,08 tok/s TG Prefill 50 · TTFT 189.477 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 31 | OpenReasoning-Nemotron-32B32BNVIDIA PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 10,99 tok/s TG Prefill 51 · TTFT 187.842 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → |