🏆
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-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 | 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 → | |
| 6 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 344,40 tok/s TG Prefill 16.771 · TTFT 2.037 ms | 10× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 123 GB RAM | tensorfoldgodclaw 1.1.5NVFP4/MXFP8 | Details → | |
| 7 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 244,85 tok/s TG Prefill 14.238 · TTFT 934 ms | 5× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 123 GB RAM | tensorfoldgodclaw 1.1.5NVFP4/MXFP8 | Details → | |
| 8 | 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 → | |
| 9 | 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 → | |
| 10 | Qwen3.8-Flash-Next180BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 74,79 tok/s TG Prefill 5.764 · TTFT 344 ms | 1× | NVIDIA RTX PRO 6000 Blackwell Workstation EditionAMD Ryzen 9 9950X 16-Core Processor · 123 GB RAM | tensorfoldgodclaw 1.1.5NVFP4/MXFP8 | Details → |