🏆
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 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 290,27 tok/s TG Prefill 35.719 · TTFT 1.015 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawBF16 | Details → | |
| 2 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 286,24 tok/s TG Prefill 2.858 · TTFT 1.792 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 3 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 277,95 tok/s TG Prefill 8.575 · TTFT 597 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 4 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 273,62 tok/s TG Prefill 5.151 · TTFT 994 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 5 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 270,40 tok/s TG Prefill 6.824 · TTFT 750 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 6 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 265,88 tok/s TG Prefill 2.862 · TTFT 894 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 7 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 260,59 tok/s TG Prefill 6.954 · TTFT 368 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 8 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 256,65 tok/s TG Prefill 4.749 · TTFT 539 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 9 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 253,13 tok/s TG Prefill 5.903 · TTFT 434 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 10 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 247,13 tok/s TG Prefill 9.886 · TTFT 518 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 11 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 244,04 tok/s TG Prefill 5.934 · TTFT 863 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 12 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 241,82 tok/s TG Prefill 7.918 · TTFT 647 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 13 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 216,36 tok/s TG Prefill 8.015 · TTFT 319 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 14 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 210,46 tok/s TG Prefill 6.833 · TTFT 375 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 15 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 202,20 tok/s TG Prefill 5.476 · TTFT 468 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 16 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 170,48 tok/s TG Prefill 23.477 · TTFT 582 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawBF16 | Details → | |
| 17 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 127,39 tok/s TG Prefill 1.511 · TTFT 3.388 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 18 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 121,30 tok/s TG Prefill 2.872 · TTFT 178 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 19 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 113,57 tok/s TG Prefill 2.783 · TTFT 184 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 20 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 108,12 tok/s TG Prefill 2.777 · TTFT 184 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 21 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 105,94 tok/s TG Prefill 2.835 · TTFT 181 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ4_K_M | Details → | |
| 22 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 103,17 tok/s TG Prefill 1.484 · TTFT 1.725 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 23 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 93,53 tok/s TG Prefill 2.671 · TTFT 192 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 24 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 91,48 tok/s TG Prefill 3.221 · TTFT 159 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 25 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 90,81 tok/s TG Prefill 3.309 · TTFT 155 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → | |
| 26 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 53,38 tok/s TG Prefill 6.812 · TTFT 427 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | vLLMgodclawBF16 | Details → | |
| 27 | Nemotron-3.5-Lightning-30B-A3B30BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 50,67 tok/s TG Prefill 1.300 · TTFT 394 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli batched-bench ROCmQ8_0 | Details → |