🏆
Rankings
LLM Leaderboard
Real measurements of every tested language model on real, named hardware – rated by raw speed (performance in tokens per second, prefill and time to first token) and by practical task quality in complete agent and chat runs (harness). Pick a benchmark type below or filter by model, maker and hardware to see exactly what is tested and how the results are produced.
⚡ Performance (tok/s)🤖 Harness quality👥 Concurrency🖥️ real hardware
| # | Model / Maker | Metrics | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 239,17 tok/s TG Prefill 2.414 · TTFT 43.060 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 203,75 tok/s TG Prefill 3.332 · TTFT 47.827 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 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 202,24 tok/s TG Prefill 2.746 · TTFT 49.827 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 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 168,46 tok/s TG Prefill 1.711 · TTFT 62.671 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 → | |
| 5 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 123,74 tok/s TG Prefill 2.668 · TTFT 13.115 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 6 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 107,05 tok/s TG Prefill 3.476 · TTFT 13.786 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 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 104,63 tok/s TG Prefill 2.633 · TTFT 14.998 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 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 86,86 tok/s TG Prefill 1.631 · TTFT 19.862 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 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 30,80 tok/s TG Prefill 1.743 · TTFT 1.412 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 10 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 29,55 tok/s TG Prefill 2.137 · TTFT 1.149 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 → | |
| 11 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 29,44 tok/s TG Prefill 1.610 · TTFT 1.524 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 → | |
| 12 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 25,49 tok/s TG Prefill 1.149 · TTFT 2.146 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 → | |
| 13 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 5,78 tok/s TG Prefill 173 · TTFT 53.622 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 14 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 5,77 tok/s TG Prefill 156 · TTFT 56.974 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 15 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 4,90 tok/s TG Prefill 239 · TTFT 40.388 ms | 5× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketQ4_K_M | Details → | |
| 16 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 4,84 tok/s TG Prefill 239 · TTFT 40.453 ms | 10× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketQ4_K_M | Details → | |
| 17 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 2,96 tok/s TG Prefill 158 · TTFT 53.509 ms | 5× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → | |
| 18 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 2,83 tok/s TG Prefill 166 · TTFT 56.632 ms | 10× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → | |
| 19 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 2,52 tok/s TG Prefill 148 · TTFT 16.554 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 20 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkPerformancetest Small 1.0 | 2,28 tok/s TG Prefill 81 · TTFT 569 ms | 1× | 2x NVIDIA GeForce RTX 20604x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz · 504 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 21 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmark | 2,28 tok/s TG Prefill 81 · TTFT 569 ms | 1× | Keine GPU (CPU-only)51x Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz · 504 GB RAM | godclaw | Details → | |
| 22 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,81 tok/s TG Prefill 15 · TTFT 141.412 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 23 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,79 tok/s TG Prefill 15 · TTFT 143.151 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 24 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,78 tok/s TG Prefill 15 · TTFT 143.025 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 25 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,76 tok/s TG Prefill 20 · TTFT 393.908 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 26 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,72 tok/s TG Prefill 21 · TTFT 372.410 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 27 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,72 tok/s TG Prefill 21 · TTFT 376.885 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 28 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,71 tok/s TG Prefill 21 · TTFT 376.282 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 29 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,70 tok/s TG Prefill 21 · TTFT 377.119 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 30 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,63 tok/s TG Prefill 21 · TTFT 377.401 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 31 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,56 tok/s TG Prefill 151 · TTFT 14.186 ms | 1× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketQ4_K_M | Details → | |
| 32 | Qwen2.5-72B-Instruct72BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 0,82 tok/s TG Prefill 116 · TTFT 18.256 ms | 1× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → |