🏆
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 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 20,60 tok/s TG Prefill 168 · TTFT 72.538 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 2 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 20,05 tok/s TG Prefill 165 · TTFT 64.548 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 3 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 18,92 tok/s TG Prefill 218 · TTFT 111.123 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 4 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 17,86 tok/s TG Prefill 166 · TTFT 63.926 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 5 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 17,26 tok/s TG Prefill 218 · TTFT 111.171 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 6 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 16,91 tok/s TG Prefill 161 · TTFT 66.087 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 7 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 15,90 tok/s TG Prefill 215 · TTFT 113.000 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 8 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 15,48 tok/s TG Prefill 220 · TTFT 110.219 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 9 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 14,49 tok/s TG Prefill 162 · TTFT 65.453 ms | 5× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 10 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 13,01 tok/s TG Prefill 118 · TTFT 17.314 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 11 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,81 tok/s TG Prefill 119 · TTFT 17.125 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 12 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,65 tok/s TG Prefill 120 · TTFT 16.976 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 13 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 12,56 tok/s TG Prefill 116 · TTFT 17.650 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 14 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,52 tok/s TG Prefill 132 · TTFT 17.401 ms | 1× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → | |
| 15 | DeepSeek-V4-Flash290.9BDeepSeek Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 11,48 tok/s TG Prefill 207 · TTFT 124.898 ms | 10× | 4x AMD Radeon AI PRO R9700AMD Ryzen Threadripper PRO 7955WX 16-Cores · 184 GB RAM | llama.cppgodclawUD-Q4_K_XL | Details → |