🏆
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 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 637,17 tok/s TG Prefill 218 · TTFT 7.760 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 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 628,63 tok/s TG Prefill 148 · TTFT 9.358 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 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 612,90 tok/s TG Prefill 149 · TTFT 8.949 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 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 534,84 tok/s TG Prefill 10.455 · TTFT 3.204 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 5 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 524,66 tok/s TG Prefill 239 · TTFT 10.174 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 → | |
| 6 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 492,95 tok/s TG Prefill 9.614 · TTFT 3.459 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 7 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 334,04 tok/s TG Prefill 103 · TTFT 2.521 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 → | |
| 8 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 332,52 tok/s TG Prefill 146 · TTFT 1.828 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 → | |
| 9 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 327,23 tok/s TG Prefill 8.158 · TTFT 1.764 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 10 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 323,37 tok/s TG Prefill 186 · TTFT 2.106 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 → | |
| 11 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 306,07 tok/s TG Prefill 7.668 · TTFT 1.898 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 12 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 278,10 tok/s TG Prefill 5.436 · TTFT 5.880 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliAWQ | Details → | |
| 13 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 258,91 tok/s TG Prefill 357 · TTFT 2.070 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 → | |
| 14 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 229,95 tok/s TG Prefill 655 · TTFT 21.862 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 15 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 224,79 tok/s TG Prefill 606 · TTFT 26.540 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 16 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 210,79 tok/s TG Prefill 606 · TTFT 28.206 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 17 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 207,60 tok/s TG Prefill 2.947 · TTFT 16.548 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 18 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 207,02 tok/s TG Prefill 2.897 · TTFT 16.703 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 19 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 194,44 tok/s TG Prefill 2.527 · TTFT 19.500 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 20 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 192,56 tok/s TG Prefill 2.801 · TTFT 18.904 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 21 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 168,52 tok/s TG Prefill 4.154 · TTFT 3.278 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliAWQ | Details → | |
| 22 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 164,07 tok/s TG Prefill 2.080 · TTFT 11.200 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 23 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 126,89 tok/s TG Prefill 2.125 · TTFT 10.798 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 24 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 118,51 tok/s TG Prefill 482 · TTFT 2.854 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 25 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 113,19 tok/s TG Prefill 2.335 · TTFT 4.870 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 26 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 112,45 tok/s TG Prefill 2.423 · TTFT 4.657 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 27 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 112,02 tok/s TG Prefill 553 · TTFT 3.143 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 28 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 111,95 tok/s TG Prefill 473 · TTFT 3.796 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 29 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 96,97 tok/s TG Prefill 28 · TTFT 287 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 → | |
| 30 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 96,44 tok/s TG Prefill 1.844 · TTFT 6.131 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 31 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 96,10 tok/s TG Prefill 30 · TTFT 268 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 → | |
| 32 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 95,86 tok/s TG Prefill 54 · TTFT 196 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 → | |
| 33 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 84,92 tok/s TG Prefill 136 · TTFT 59 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 → | |
| 34 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 83,06 tok/s TG Prefill 2.594 · TTFT 4.350 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 35 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 83,01 tok/s TG Prefill 2.572 · TTFT 4.388 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 36 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 79,48 tok/s TG Prefill 3.787 · TTFT 566 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 37 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 78,65 tok/s TG Prefill 1.940 · TTFT 5.826 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 38 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 75,04 tok/s TG Prefill 3.417 · TTFT 628 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawAWQ | Details → | |
| 39 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 44,45 tok/s TG Prefill 1.820 · TTFT 1.238 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 40 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 44,41 tok/s TG Prefill 1.819 · TTFT 1.240 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 41 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 40,37 tok/s TG Prefill 2.099 · TTFT 1.022 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMopenclaw_cliAWQ | Details → | |
| 42 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 38,40 tok/s TG Prefill 1.782 · TTFT 1.265 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 43 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 34,29 tok/s TG Prefill 251 · TTFT 32 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 44 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 34,25 tok/s TG Prefill 254 · TTFT 31 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 45 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 33,92 tok/s TG Prefill 252 · TTFT 32 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 46 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 28,90 tok/s TG Prefill 1.318 · TTFT 2.246 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 47 | Magistral-Small-250924BMistral AI Performancebenchmark | 28,90 tok/s TG Prefill 1.318 · TTFT 2.246 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | godclawQ4_K_M | Details → | |
| 48 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 27,20 tok/s TG Prefill 1.906 · TTFT 1.183 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 49 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 27,20 tok/s TG Prefill 1.896 · TTFT 1.189 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ8_0 | Details → | |
| 50 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 23,81 tok/s TG Prefill 1.859 · TTFT 1.214 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cliQ8_0 | Details → | |
| 51 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 19,42 tok/s TG Prefill 2.175 · TTFT 1.030 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 52 | Magistral-Small-250924BMistral AI Performancebenchmark | 19,42 tok/s TG Prefill 2.175 · TTFT 1.030 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | godclawQ4_K_M | Details → | |
| 53 | Magistral-Small-250924BMistral AI PerformancebenchmarkTimebench 3 - Kombi (Prefill + Generation) | 18,46 tok/s TG Prefill 1.263 · TTFT 1.741 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 54 | Magistral-Small-250924BMistral AI Performancebenchmark | 18,46 tok/s TG Prefill 1.263 · TTFT 1.741 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | godclawQ4_K_M | Details → |