🏆 Rangliste
Highscore
Die besten Läufe im direkten Ranking – umschaltbar zwischen Performance (Token/s) und Harness (Aufgabenpunkte).
| # | Modell / Hersteller | Grafikkarte | CPU / RAM | Kombi | Prefill | TG | TTFT | Parallel | Runtime | |
|---|---|---|---|---|---|---|---|---|---|---|
| 🥇 | gemma-4-E2B-it5BGoogle | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 9.308,4 tok/s |
55.260 | 1.568,0 | 492 ms | 10× | vLLMgodclawFP8 | Details → |
| 🥈 | Nemotron-3-Nano-4B4BNVIDIA | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 8.099,4 tok/s |
44.171 | 1.485,1 | 716 ms | 10× | vLLMgodclawFP8 | Details → |
| 🥉 | gemma-4-E2B-it5BGoogle | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 8.005,8 tok/s |
47.158 | 1.359,1 | 587 ms | 10× | vLLMgodclawFP8 | Details → |
| 4 | gpt-oss-20b20BOpenAI | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 7.144,1 tok/s |
35.348 | 1.443,9 | 560 ms | 10× | vLLMgodclaw | Details → |
| 5 | gemma-4-E2B-it5BGoogle | NVIDIA GeForce RTX 5070 Ti | AMD Ryzen Threadripper PRO 5975WX 32-Cores 247 GB | 7.112,6 tok/s |
31.452 | 1.608,5 | 1632 ms | 10× | vLLMgodclaw | Details → |
| 6 | gemma-4-E2B-it5BGoogle | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 6.996,8 tok/s |
20.277 | 2.414,4 | 4106 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 7 | gemma-4-E2B-it5BGoogle | NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 6.592,0 tok/s |
38.712 | 1.122,5 | 690 ms | 10× | vLLMopenclaw_cliFP8 | Details → |
| 8 | gemma-4-E2B-it5BGoogle | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 6.276,7 tok/s |
15.814 | 2.491,2 | 4382 ms | 10× | llama.cppcodex_cliQ4_K_M | Details → |
| 9 | gemma-4-E2B-it5BGoogle | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 6.258,7 tok/s |
43.093 | 909,0 | 287 ms | 5× | vLLMgodclawFP8 | Details → |
| 10 | Qwen3-VL-30B-A3B-Instruct30BQwen (Alibaba) | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 6.159,3 tok/s |
36.884 | 1.028,5 | 915 ms | 10× | vLLMgodclawAWQ | Details → |
| 11 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 6.010,1 tok/s |
33.741 | 1.070,6 | 975 ms | 10× | vLLMgodclawAWQ | Details → |
| 12 | gpt-oss-20b20BOpenAI | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.991,9 tok/s |
19.732 | 1.819,5 | 5578 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 13 | Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.946,6 tok/s |
39.029 | 906,0 | 1192 ms | 10× | vLLMgodclawAWQ | Details → |
| 14 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 5.809,3 tok/s |
37.116 | 909,3 | 1783 ms | 10× | vLLMcodex_cliAWQ | Details → |
| 15 | gpt-oss-20b20BOpenAI | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 5.742,9 tok/s |
25.462 | 1.295,3 | 808 ms | 10× | vLLMcodex_cliAWQ | Details → |
| 16 | gpt-oss-20b20BOpenAI | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.730,8 tok/s |
18.226 | 1.802,0 | 5604 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 17 | Laguna-XS-2.1Poolside | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.722,7 tok/s |
39.157 | 836,4 | 1032 ms | 10× | vLLMgodclawAWQ | Details → |
| 18 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 5.712,0 tok/s |
33.768 | 966,2 | 1792 ms | 10× | vLLMcodex_cliAWQ | Details → |
| 19 | gpt-oss-20b20BOpenAI | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.703,6 tok/s |
16.318 | 1.993,5 | 5322 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 20 | gpt-oss-20b20BOpenAI | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.678,2 tok/s |
16.158 | 1.995,4 | 5309 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 21 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.573,3 tok/s |
32.305 | 961,5 | 1633 ms | 10× | vLLMgodclawAWQ | Details → |
| 22 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.573,0 tok/s |
20.491 | 1.515,7 | 6859 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 23 | Qwen3-Omni-30B-A3B-Thinking30BQwen (Alibaba) | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.557,0 tok/s |
33.907 | 910,7 | 1527 ms | 10× | vLLMgodclawAWQ | Details → |
| 24 | Meta-Llama-3.1-8B-Instruct8BMeta | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.487,2 tok/s |
21.012 | 1.433,0 | 6891 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 25 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.467,7 tok/s |
19.346 | 1.545,3 | 6758 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 26 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.453,0 tok/s |
37.676 | 789,2 | 952 ms | 10× | vLLMgodclawAWQ | Details → |
| 27 | gpt-oss-20b20BOpenAI | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.414,9 tok/s |
15.972 | 1.835,7 | 5760 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 28 | gemma-4-E2B-it5BGoogle | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.405,6 tok/s |
13.632 | 2.143,6 | 4819 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 29 | Qwen3-VL-30B-A3B-Instruct30BQwen (Alibaba) | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.388,4 tok/s |
19.292 | 1.505,0 | 6959 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 30 | Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 5.371,5 tok/s |
33.023 | 873,7 | 1834 ms | 10× | vLLMcodex_cliAWQ | Details → |
| 31 | gemma-4-E2B-it5BGoogle | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.359,1 tok/s |
35.757 | 803,2 | 347 ms | 5× | vLLMgodclawFP8 | Details → |
| 32 | gpt-oss-20b20BOpenAI | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 5.284,6 tok/s |
14.771 | 1.890,7 | 5489 ms | 10× | llama.cppcodex_cliQ4_K_M | Details → |
| 33 | gpt-oss-20b20BOpenAI | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 5.264,0 tok/s |
19.981 | 1.386,8 | 1205 ms | 10× | vLLMcodex_cli | Details → |
| 34 | Meta-Llama-3.1-8B-Instruct8BMeta | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.245,4 tok/s |
26.805 | 1.026,4 | 1370 ms | 10× | vLLMgodclawAWQ | Details → |
| 35 | Qwen3-VL-30B-A3B-Instruct30BQwen (Alibaba) | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.235,4 tok/s |
35.900 | 763,5 | 986 ms | 10× | vLLMgodclawAWQ | Details → |
| 36 | Meta-Llama-3.1-8B-Instruct8BMeta | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.232,5 tok/s |
17.861 | 1.533,0 | 6224 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 37 | Nemotron-3-Nano-4B4BNVIDIA | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.213,0 tok/s |
30.214 | 899,4 | 472 ms | 5× | vLLMgodclawFP8 | Details → |
| 38 | gpt-oss-20b20BOpenAI | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.200,7 tok/s |
21.843 | 1.238,3 | 905 ms | 10× | vLLMgodclaw | Details → |
| 39 | Gemma-4-26B-A4B-it26BGoogle | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.192,2 tok/s |
32.027 | 841,8 | 922 ms | 10× | vLLMgodclawAWQ | Details → |
| 40 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.146,7 tok/s |
34.913 | 758,7 | 991 ms | 10× | vLLMgodclawAWQ | Details → |
| 41 | Laguna-XS-2.1Poolside | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.137,2 tok/s |
34.194 | 771,8 | 1183 ms | 10× | vLLMgodclawAWQ | Details → |
| 42 | Qwen3-Coder-30B-A3B-Instruct30BQwen (Alibaba) | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 5.121,0 tok/s |
17.119 | 1.531,9 | 6936 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
| 43 | gpt-oss-20b20BOpenAI | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.119,9 tok/s |
31.098 | 842,9 | 318 ms | 5× | vLLMgodclaw | Details → |
| 44 | Meta-Llama-3.1-8B-Instruct8BMeta | NVIDIA GeForce RTX 5090 | AMD Ryzen 7 5800X3D 8-Core Processor 126 GB | 5.107,7 tok/s |
16.376 | 1.593,1 | 6848 ms | 10× | llama.cppcodex_cliQ4_K_M | Details → |
| 45 | Qwen3.6-35B-A3B35BQwen (Alibaba) | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.085,6 tok/s |
32.313 | 800,4 | 826 ms | 10× | vLLMgodclawAWQ | Details → |
| 46 | Qwen3.5-35B-A3B35BQwen (Alibaba) | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.059,9 tok/s |
32.522 | 787,3 | 826 ms | 10× | vLLMgodclawAWQ | Details → |
| 47 | Qwen-AgentWorld-35B-A3B35BQwen (Alibaba) | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.054,9 tok/s |
31.954 | 799,7 | 833 ms | 10× | vLLMgodclawAWQ | Details → |
| 48 | Qwen3.5-35B-A3B35BQwen (Alibaba) | NVIDIA RTX PRO 6000 Blackwell Workstation Edition | AMD Ryzen 9 9950X 16-Core Processor 92 GB | 5.016,0 tok/s |
24.461 | 1.028,6 | 1670 ms | 10× | vLLMgodclawAWQ | Details → |
| 49 | Laguna-XS-2.1Poolside | 2x NVIDIA RTX A6000 | AMD Ryzen Threadripper PRO 7955WX 16-Cores 184 GB | 5.002,3 tok/s |
35.143 | 712,0 | 1137 ms | 10× | vLLMgodclawFP8 | Details → |
| 50 | Qwen3-30B-A3B-Instruct-250730BQwen (Alibaba) | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition | AMD Ryzen Threadripper PRO 9965WX 24-Cores 125 GB | 4.974,0 tok/s |
16.412 | 1.507,5 | 7064 ms | 10× | llama.cppgodclawQ4_K_M | Details → |
Ranking aus veröffentlichten Benchmarks. Sortiert nach Generation-Durchsatz. Wie wird gemessen?
