🏆
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 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B120BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.562,27 tok/s TG Prefill 2.593 · TTFT 34.854 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 2 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B120BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 795,03 tok/s TG Prefill 2.378 · TTFT 15.951 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 3 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B120BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 158,32 tok/s TG Prefill 1.225 · TTFT 2.121 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition iAMD Ryzen 9 7945HX with Radeon Graphics · 60 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 4 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B120BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 79,84 tok/s TG Prefill 3.760 · TTFT 5.878 ms | 10× | NVIDIA GB10 (DGX Spark) iNVIDIA Grace · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 5 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B120BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 54,14 tok/s TG Prefill 3.027 · TTFT 3.561 ms | 5× | NVIDIA GB10 (DGX Spark) iNVIDIA Grace · 120 GB RAM | vLLMgodclawNVFP4 | Details → | |
| 6 | NVIDIA-Nemotron-3.5-Super-VL-120B-A12B120BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 20,44 tok/s TG Prefill 1.717 · TTFT 1.294 ms | 1× | NVIDIA GB10 (DGX Spark) iNVIDIA Grace · 120 GB RAM | vLLMgodclawNVFP4 | Details → |