⚡
Performance benchmark
How fast is the model?
Raw throughput on real hardware – how many tokens a model generates per second, how quickly it processes the prompt and how short the time to first token is. Higher is better.
⚡ Generation (tok/s)🔄 Prefill⏱️ TTFT👥 Concurrency
| # | Model / Maker | Metrics | Parallel | GPU / CPU / RAM | Runtime | ||
|---|---|---|---|---|---|---|---|
| 1 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 3.484,32 tok/s TG Prefill 8.256 · TTFT 14.388 ms | 10× | NVIDIA GeForce RTX 3090AMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | tensorfoldgodclaw 1.1.6MLX 4-bit | Details → | |
| 2 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 2.867,15 tok/s TG Prefill 8.107 · TTFT 17.054 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | tensorfoldgodclaw 1.1.6MLX 4-bit | Details → | |
| 3 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.694,65 tok/s TG Prefill 7.768 · TTFT 6.482 ms | 5× | NVIDIA GeForce RTX 3090AMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | tensorfoldgodclaw 1.1.6MLX 4-bit | Details → | |
| 4 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.375,08 tok/s TG Prefill 7.576 · TTFT 7.731 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | tensorfoldgodclaw 1.1.6MLX 4-bit | Details → | |
| 5 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1.132,67 tok/s TG Prefill 1.930 · TTFT 47.978 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | stratagodclawIQ2_XS | Details → | |
| 6 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 725,25 tok/s TG Prefill 16.303 · TTFT 1.813 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | vLLMgodclaw 0.27.1NVFP4 W4A16 | Details → | |
| 7 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 709,02 tok/s TG Prefill 16.443 · TTFT 1.805 ms | 10× | NVIDIA GeForce RTX 3090AMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | vLLMgodclaw 0.27.1NVFP4 W4A16 | Details → | |
| 8 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 638,58 tok/s TG Prefill 306 · TTFT 145.278 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | stratagodclawIQ3_XXS | Details → | |
| 9 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 572,23 tok/s TG Prefill 1.826 · TTFT 21.491 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | stratagodclawIQ2_XS | Details → | |
| 10 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkKurzaufgaben TTFT - Thinking OFF - Outputlimit 1/5/10 V1 | 471,33 tok/s TG Prefill - · TTFT 31 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | tensorfoldgodclaw 1.1.6MLX 4-bit | Details → | |
| 11 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 466,72 tok/s TG Prefill 12.079 · TTFT 1.048 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | vLLMgodclaw 0.27.1NVFP4 W4A16 | Details → | |
| 12 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 462,42 tok/s TG Prefill 11.388 · TTFT 1.082 ms | 5× | NVIDIA GeForce RTX 3090AMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | vLLMgodclaw 0.27.1NVFP4 W4A16 | Details → | |
| 13 | gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 448,17 tok/s TG Prefill 5.809 · TTFT 3.494 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 14 | Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 371,66 tok/s TG Prefill 8.745 · TTFT 3.710 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 15 | gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 361,70 tok/s TG Prefill 4.756 · TTFT 2.076 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 16 | Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 341,53 tok/s TG Prefill 4.662 · TTFT 4.820 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 17 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 319,40 tok/s TG Prefill 254 · TTFT 73.516 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | stratagodclawIQ3_XXS | Details → | |
| 18 | Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 303,07 tok/s TG Prefill 6.279 · TTFT 2.200 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 19 | Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 299,90 tok/s TG Prefill 3.515 · TTFT 2.858 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 20 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 235,73 tok/s TG Prefill 6.703 · TTFT 322 ms | 1× | NVIDIA GeForce RTX 3090AMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | tensorfoldgodclaw 1.1.6MLX 4-bit | Details → | |
| 21 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkKurzaufgaben TTFT - Thinking OFF - Outputlimit 1/5/10 V1 | 218,77 tok/s TG Prefill - · TTFT 54 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | vLLMgodclaw 0.27.1NVFP4 W4A16 | Details → | |
| 22 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 211,38 tok/s TG Prefill 5.434 · TTFT 411 ms | 1× | NVIDIA GeForce RTX 3090AMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | vLLMgodclaw 0.27.1NVFP4 W4A16 | Details → | |
| 23 | gpt-oss-20b20BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 211,09 tok/s TG Prefill 4.232 · TTFT 467 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 24 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 210,66 tok/s TG Prefill 6.076 · TTFT 355 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | vLLMgodclaw 0.27.1NVFP4 W4A16 | Details → | |
| 25 | Qwen3-30B-A3B-Thinking-250730BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 189,23 tok/s TG Prefill 3.806 · TTFT 640 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 26 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 175,99 tok/s TG Prefill 6.612 · TTFT 326 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | tensorfoldgodclaw 1.1.6MLX 4-bit | Details → | |
| 27 | Qwen3.6-35B-A3B35BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 172,41 tok/s TG Prefill 2.763 · TTFT 700 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 28 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 139,97 tok/s TG Prefill 1.744 · TTFT 14.686 ms | 10× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 29 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 119,24 tok/s TG Prefill 1.181 · TTFT 9.551 ms | 5× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 30 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 116,39 tok/s TG Prefill 1.389 · TTFT 1.421 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | stratagodclawIQ2_XS | Details → | |
| 31 | NVIDIA-Nemotron-3.5-Lightning-30B-A3B31.6BNVIDIA Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 101,04 tok/s TG Prefill 1.039 · TTFT 2.074 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 32 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 50,53 tok/s TG Prefill 106 · TTFT 20.413 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | stratagodclawIQ3_XXS | Details → | |
| 33 | gpt-oss-120b120BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 18,07 tok/s TG Prefill 96 · TTFT 20.531 ms | 1× | NVIDIA GeForce RTX 3090 iAMD Ryzen 5 5600X 6-Core Processor · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → |