🏆
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 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 835,13 tok/s TG Prefill 3.499 · TTFT 16.577 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 2 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 797,14 tok/s TG Prefill 3.616 · TTFT 16.087 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 3 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 791,09 tok/s TG Prefill 3.559 · TTFT 16.497 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 4 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 781,43 tok/s TG Prefill 3.666 · TTFT 16.232 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawINT4 | Details → | |
| 5 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 779,78 tok/s TG Prefill 3.841 · TTFT 16.300 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 6 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 777,69 tok/s TG Prefill 3.903 · TTFT 16.458 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 7 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 776,08 tok/s TG Prefill 4.515 · TTFT 15.579 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 8 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 774,90 tok/s TG Prefill 3.820 · TTFT 15.880 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 9 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 774,10 tok/s TG Prefill 3.866 · TTFT 16.399 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 10 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 762,72 tok/s TG Prefill 4.616 · TTFT 15.283 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 11 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 706,85 tok/s TG Prefill 2.752 · TTFT 19.766 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 12 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 697,82 tok/s TG Prefill 2.814 · TTFT 19.437 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 13 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 662,97 tok/s TG Prefill 2.731 · TTFT 19.776 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 14 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 470,52 tok/s TG Prefill 3.067 · TTFT 6.420 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 15 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 465,82 tok/s TG Prefill 3.019 · TTFT 6.496 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 16 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 461,99 tok/s TG Prefill 3.219 · TTFT 6.266 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 17 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 456,52 tok/s TG Prefill 3.165 · TTFT 6.386 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawINT4 | Details → | |
| 18 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 440,78 tok/s TG Prefill 3.886 · TTFT 5.752 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 19 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 437,16 tok/s TG Prefill 3.329 · TTFT 6.298 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 20 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 434,94 tok/s TG Prefill 3.365 · TTFT 6.267 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 21 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 434,78 tok/s TG Prefill 3.944 · TTFT 5.525 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 22 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 429,63 tok/s TG Prefill 3.334 · TTFT 6.284 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 23 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 415,35 tok/s TG Prefill 3.365 · TTFT 5.876 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 24 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 413,18 tok/s TG Prefill 2.358 · TTFT 8.016 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 25 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 408,67 tok/s TG Prefill 2.354 · TTFT 8.049 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 26 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 405,17 tok/s TG Prefill 2.414 · TTFT 7.850 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 27 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 356,42 tok/s TG Prefill 16.357 · TTFT 2.480 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 28 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 344,15 tok/s TG Prefill 15.326 · TTFT 2.709 ms | 10× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 29 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 250,44 tok/s TG Prefill 13.079 · TTFT 1.402 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 30 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 234,43 tok/s TG Prefill 11.921 · TTFT 1.518 ms | 5× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 31 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 146,11 tok/s TG Prefill 2.580 · TTFT 1.064 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 32 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 145,95 tok/s TG Prefill 2.846 · TTFT 964 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclawINT4 | Details → | |
| 33 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 135,75 tok/s TG Prefill 3.130 · TTFT 877 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 34 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 135,51 tok/s TG Prefill 2.905 · TTFT 945 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 35 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 135,22 tok/s TG Prefill 2.941 · TTFT 933 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 36 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 135,13 tok/s TG Prefill 2.928 · TTFT 938 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 37 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 133,97 tok/s TG Prefill 2.205 · TTFT 1.246 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawINT4 | Details → | |
| 38 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 131,71 tok/s TG Prefill 2.197 · TTFT 1.250 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 39 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 109,52 tok/s TG Prefill 5.943 · TTFT 462 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 40 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 100,14 tok/s TG Prefill 4.429 · TTFT 645 ms | 1× | 2x NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 41 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 70,51 tok/s TG Prefill 5.970 · TTFT 7.475 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 42 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 42,33 tok/s TG Prefill 153 · TTFT 257.193 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 43 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 42,20 tok/s TG Prefill 4.890 · TTFT 4.010 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 44 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,82 tok/s TG Prefill 177 · TTFT 255.495 ms | 10× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketINT4 | Details → | |
| 45 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,70 tok/s TG Prefill 134 · TTFT 130.995 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 46 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,47 tok/s TG Prefill 97 · TTFT 309.056 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 47 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,27 tok/s TG Prefill 100 · TTFT 306.428 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 48 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 31,55 tok/s TG Prefill 174 · TTFT 246.888 ms | 10× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketUD-Q4_K_M | Details → | |
| 49 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 31,51 tok/s TG Prefill 98 · TTFT 302.732 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 50 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 29,83 tok/s TG Prefill 87 · TTFT 185.676 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 51 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 29,65 tok/s TG Prefill 86 · TTFT 187.895 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 52 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 28,64 tok/s TG Prefill 89 · TTFT 183.222 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 53 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 25,54 tok/s TG Prefill 157 · TTFT 125.314 ms | 5× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketUD-Q4_K_M | Details → | |
| 54 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 25,50 tok/s TG Prefill 157 · TTFT 125.739 ms | 5× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketINT4 | Details → | |
| 55 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 13,37 tok/s TG Prefill 8.629 · TTFT 305 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawINT4 | Details → | |
| 56 | Laguna-S-2.1-INT4Poolside Performance benchmark | 13,37 tok/s TG Prefill 8.629 · TTFT 305 ms | 1× | 3x AMD Radeon AI PRO R970032x AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | godclawINT4 | Details → | |
| 57 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 13,32 tok/s TG Prefill 98 · TTFT 206.915 ms | 5× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliUD-Q4_K_M | Details → | |
| 58 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,62 tok/s TG Prefill 130 · TTFT 21.118 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 59 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,64 tok/s TG Prefill 76 · TTFT 36.235 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 60 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,60 tok/s TG Prefill 76 · TTFT 36.406 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 61 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,44 tok/s TG Prefill 76 · TTFT 36.141 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_M | Details → | |
| 62 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 9,87 tok/s TG Prefill 88 · TTFT 196.186 ms | 10× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliUD-Q4_K_M | Details → | |
| 63 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 9,12 tok/s TG Prefill 157 · TTFT 17.486 ms | 1× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketINT4 | Details → | |
| 64 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 9,06 tok/s TG Prefill 156 · TTFT 17.644 ms | 1× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocketUD-Q4_K_M | Details → | |
| 65 | Laguna-S-2.1-INT4Poolside Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 4,80 tok/s TG Prefill 100 · TTFT 27.502 ms | 1× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cliUD-Q4_K_M | Details → |