🏆
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 | gpt-oss-120b120BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 135,08 tok/s TG Prefill 1.548 · TTFT 12.948 ms | 10× | NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | gpt-oss-120b120BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 100,67 tok/s TG Prefill 1.446 · TTFT 6.855 ms | 5× | NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | gpt-oss-120b120BOpenAI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 62,23 tok/s TG Prefill 1.354 · TTFT 1.459 ms | 1× | NVIDIA DGX SPARK (GB10) iNVIDIA Grace · 120 GB RAM | llama.cppgodclawQ4_K_M | Details → |

