🏆
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 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 141,57 tok/s TG Prefill 2.054 · TTFT 71.488 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 73,58 tok/s TG Prefill 1.877 · TTFT 22.218 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 19,65 tok/s TG Prefill 1.170 · TTFT 2.148 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 4 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 5,77 tok/s TG Prefill 334 · TTFT 5.648 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 5 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmark | 5,77 tok/s TG Prefill 334 · TTFT 5.648 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | godclaw | Details → | |
| 6 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 4,16 tok/s TG Prefill 149 · TTFT 69.107 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 7 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 4,14 tok/s TG Prefill 150 · TTFT 68.889 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 8 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 2,83 tok/s TG Prefill 109 · TTFT 85.257 ms | 5× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketQ4_K_M | Details → | |
| 9 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 2,74 tok/s TG Prefill 145 · TTFT 71.210 ms | 10× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketQ4_K_M | Details → | |
| 10 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,93 tok/s TG Prefill 128 · TTFT 80.836 ms | 10× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → | |
| 11 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,93 tok/s TG Prefill 127 · TTFT 81.345 ms | 10× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → | |
| 12 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,90 tok/s TG Prefill 128 · TTFT 80.794 ms | 5× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → | |
| 13 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,86 tok/s TG Prefill 99 · TTFT 93.863 ms | 5× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → | |
| 14 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,71 tok/s TG Prefill 95 · TTFT 22.448 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 15 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,37 tok/s TG Prefill 10 · TTFT 221.876 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 16 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,35 tok/s TG Prefill 10 · TTFT 215.803 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 17 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,34 tok/s TG Prefill 10 · TTFT 222.230 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 18 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,34 tok/s TG Prefill 10 · TTFT 222.874 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 19 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,33 tok/s TG Prefill 10 · TTFT 221.753 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 20 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,33 tok/s TG Prefill 10 · TTFT 222.221 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 21 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,32 tok/s TG Prefill 10 · TTFT 221.650 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 22 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,32 tok/s TG Prefill 9 · TTFT 224.878 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 23 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,31 tok/s TG Prefill 10 · TTFT 216.429 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 24 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,29 tok/s TG Prefill 10 · TTFT 219.851 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 25 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,28 tok/s TG Prefill 10 · TTFT 216.526 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 26 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,27 tok/s TG Prefill 10 · TTFT 216.964 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 27 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 0,88 tok/s TG Prefill 94 · TTFT 22.520 ms | 1× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketQ4_K_M | Details → | |
| 28 | Mistral-Medium-3.5-128B128BMistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 0,56 tok/s TG Prefill 84 · TTFT 25.429 ms | 1× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliQ4_K_M | Details → |