🏆
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-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.768,14 tok/s TG Prefill 8.360 · TTFT 6.952 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclaw | Details → | |
| 2 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.719,17 tok/s TG Prefill 8.044 · TTFT 7.376 ms | 10× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cli | Details → | |
| 3 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.655,21 tok/s TG Prefill 10.591 · TTFT 7.004 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 4 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.650,72 tok/s TG Prefill 11.398 · TTFT 6.887 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 5 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.542,64 tok/s TG Prefill 7.398 · TTFT 8.118 ms | 10× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 6 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.007,26 tok/s TG Prefill 7.851 · TTFT 2.444 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclaw | Details → | |
| 7 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 986,08 tok/s TG Prefill 6.401 · TTFT 2.783 ms | 5× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cli | Details → | |
| 8 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 928,51 tok/s TG Prefill 10.134 · TTFT 2.295 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 9 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 926,98 tok/s TG Prefill 9.358 · TTFT 2.375 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 10 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 899,18 tok/s TG Prefill 6.644 · TTFT 2.846 ms | 5× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 11 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 897,47 tok/s TG Prefill 4.560 · TTFT 13.829 ms | 10× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocket | Details → | |
| 12 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 548,84 tok/s TG Prefill 4.012 · TTFT 4.835 ms | 5× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocket | Details → | |
| 13 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 359,03 tok/s TG Prefill 3.976 · TTFT 549 ms | 1× | NVIDIA GeForce RTX 5090AMD Ryzen 7 5800X3D 8-Core Processor · 126 GB RAM | llama.cppcodex_cli | Details → | |
| 14 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 356,78 tok/s TG Prefill 4.249 · TTFT 510 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 92 GB RAM | llama.cppgodclaw | Details → | |
| 15 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 330,17 tok/s TG Prefill 5.181 · TTFT 419 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 16 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 329,90 tok/s TG Prefill 6.288 · TTFT 345 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 17 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 327,65 tok/s TG Prefill 6.467 · TTFT 336 ms | 1× | 3x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation EditionAMD Ryzen Threadripper PRO 9965WX 24-Cores · 125 GB RAM | llama.cppgodclaw | Details → | |
| 18 | Nemotron-Cascade-2-30B-A3B30BBF16NVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 326,85 tok/s TG Prefill 125 · TTFT 210.255 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | vLLMgodclaw | Details → | |
| 19 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 262,44 tok/s TG Prefill 822 · TTFT 55.319 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclaw | Details → | |
| 20 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 250,32 tok/s TG Prefill 5.865 · TTFT 3.782 ms | 10× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli | Details → | |
| 21 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 221,80 tok/s TG Prefill 3.314 · TTFT 655 ms | 1× | NVIDIA GeForce RTX 3090 TiAMD Ryzen 9 8945HX with Radeon Graphics · 92 GB RAM | llama.cppwebsocket | Details → | |
| 22 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 219,79 tok/s TG Prefill 4.791 · TTFT 2.262 ms | 5× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli | Details → | |
| 23 | Nemotron-Cascade-2-30B-A3B30BIQ4_XSNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 219,59 tok/s TG Prefill 2.273 · TTFT 11.329 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen 9 7945HX with Radeon Graphics · 61 GB RAM | llama.cppgodclaw | Details → | |
| 24 | Nemotron-Cascade-2-30B-A3B30BIQ4_XSNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 190,25 tok/s TG Prefill 1.859 · TTFT 13.867 ms | 10× | NVIDIA GeForce RTX 5070 TiAMD Ryzen 9 7945HX with Radeon Graphics · 61 GB RAM | llama.cppgodclaw | Details → | |
| 25 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 162,55 tok/s TG Prefill 715 · TTFT 21.986 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclaw | Details → | |
| 26 | Nemotron-Cascade-2-30B-A3B30BIQ4_XSNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 157,50 tok/s TG Prefill 1.715 · TTFT 6.606 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen 9 7945HX with Radeon Graphics · 61 GB RAM | llama.cppgodclaw | Details → | |
| 27 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 157,01 tok/s TG Prefill 4.273 · TTFT 508 ms | 1× | NVIDIA RTX A6000AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppopenclaw_cli | Details → | |
| 28 | Nemotron-Cascade-2-30B-A3B30BIQ4_XSNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 156,26 tok/s TG Prefill 1.343 · TTFT 8.459 ms | 5× | NVIDIA GeForce RTX 5070 TiAMD Ryzen 9 7945HX with Radeon Graphics · 61 GB RAM | llama.cppgodclaw | Details → | |
| 29 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 148,79 tok/s TG Prefill 3.362 · TTFT 6.588 ms | 10× | NVIDIA GB10 (DGX Spark)NVIDIA Grace · 120 GB RAM | llama.cppopenclaw_cli | Details → | |
| 30 | Nemotron-Cascade-2-30B-A3B30BIQ4_XSNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 138,75 tok/s TG Prefill 1.439 · TTFT 1.508 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen 9 7945HX with Radeon Graphics · 61 GB RAM | llama.cppgodclaw | Details → | |
| 31 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 123,54 tok/s TG Prefill 2.897 · TTFT 3.743 ms | 5× | NVIDIA GB10 (DGX Spark)NVIDIA Grace · 120 GB RAM | llama.cppopenclaw_cli | Details → | |
| 32 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 122,13 tok/s TG Prefill 269 · TTFT 141.200 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 33 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 122,08 tok/s TG Prefill 272 · TTFT 140.873 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 34 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 121,28 tok/s TG Prefill 271 · TTFT 141.620 ms | 10× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 35 | Nemotron-Cascade-2-30B-A3B30BIQ4_XSNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 120,01 tok/s TG Prefill 1.155 · TTFT 1.875 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen 9 7945HX with Radeon Graphics · 61 GB RAM | llama.cppgodclaw | Details → | |
| 36 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) i | 107,87 tok/s TG Prefill 3.141 · TTFT 690 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 37 | Nemotron-Cascade-2-30B-A3B30BQ8_0NVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 101,36 tok/s TG Prefill 3.237 · TTFT 6.850 ms | 10× | NVIDIA GB10 (DGX Spark)NVIDIA Grace · 120 GB RAM | llama.cppopenclaw_cli | Details → | |
| 38 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 94,63 tok/s TG Prefill 1.463 · TTFT 15.382 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclaw | Details → | |
| 39 | Nemotron-Cascade-2-30B-A3B30BQ8_0NVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 84,61 tok/s TG Prefill 2.849 · TTFT 3.806 ms | 5× | NVIDIA GB10 (DGX Spark)NVIDIA Grace · 120 GB RAM | llama.cppopenclaw_cli | Details → | |
| 40 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 76,95 tok/s TG Prefill 2.356 · TTFT 929 ms | 1× | NVIDIA GB10 (DGX Spark)NVIDIA Grace · 120 GB RAM | llama.cppopenclaw_cli | Details → | |
| 41 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 76,18 tok/s TG Prefill 221 · TTFT 62.327 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 42 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 75,16 tok/s TG Prefill 220 · TTFT 62.756 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 43 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 74,62 tok/s TG Prefill 219 · TTFT 62.998 ms | 5× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 44 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 68,96 tok/s TG Prefill 623 · TTFT 3.485 ms | 1× | NVIDIA GeForce RTX 5070 TiAMD Ryzen Threadripper PRO 5975WX 32-Cores · 247 GB RAM | llama.cppgodclaw | Details → | |
| 45 | Nemotron-Cascade-2-30B-A3B30BQ8_0NVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 62,04 tok/s TG Prefill 2.341 · TTFT 935 ms | 1× | NVIDIA GB10 (DGX Spark)NVIDIA Grace · 120 GB RAM | llama.cppopenclaw_cli | Details → | |
| 46 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,17 tok/s TG Prefill 179 · TTFT 12.138 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 47 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 31,81 tok/s TG Prefill 179 · TTFT 12.150 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 48 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 31,70 tok/s TG Prefill 179 · TTFT 12.121 ms | 1× | 3x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclaw | Details → | |
| 49 | Nemotron-Cascade-2-30B-A3B30BBF16NVIDIA Performance benchmarkPerformancetest Small 1.0 | 27,50 tok/s TG Prefill 299 · TTFT 224 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | vLLMgodclaw | Details → | |
| 50 | Nemotron-Cascade-2-30B-A3B30BBF16NVIDIA Performance benchmarkPerformancetest Small 1.0 | 27,50 tok/s TG Prefill 109 · TTFT 604 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | vLLMgodclaw | Details → | |
| 51 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 5,86 tok/s TG Prefill 61 · TTFT 35.481 ms | 1× | Keine GPU (CPU-only)4x Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz · 504 GB RAM | llama.cppgodclaw | Details → | |
| 52 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 5,33 tok/s TG Prefill 74 · TTFT 150.609 ms | 5× | Keine GPU (CPU-only)4x Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz · 504 GB RAM | llama.cppgodclaw | Details → | |
| 53 | Nemotron-Cascade-2-30B-A3B30BQ4_K_MNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 0,27 tok/s TG Prefill 82 · TTFT 204.776 ms | 10× | Keine GPU (CPU-only)4x Intel(R) Xeon(R) CPU E5-4620 v2 @ 2.60GHz · 504 GB RAM | llama.cppgodclaw | Details → |