🏆
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 | Mistral-Small-24B-Instruct-2501Mistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 153,12 tok/s TG Prefill 4.348 · TTFT 7.017 ms | 10× | 2x AMD Radeon RX 7900 XTX i4x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz · 504 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | Mistral-Small-24B-Instruct-2501Mistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 86,87 tok/s TG Prefill 3.428 · TTFT 3.916 ms | 5× | 2x AMD Radeon RX 7900 XTX i4x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz · 504 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | Mistral-Small-24B-Instruct-2501Mistral AI Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 38,51 tok/s TG Prefill 1.549 · TTFT 1.390 ms | 1× | 2x AMD Radeon RX 7900 XTX i4x Intel(R) Xeon(R) CPU E5-4657L v2 @ 2.40GHz · 504 GB RAM | llama.cppgodclawQ4_K_M | Details → |