🏆
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 | SmolLM2-1.7B-Instruct1.7BHuggingFaceTB Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 531,74 tok/s TG Prefill 44.141 · TTFT 1.002 ms | 10× | 2x AMD Radeon RX 7900 XTXAMD Ryzen Threadripper PRO 3955WX 16-Cores · 63 GB RAM | vLLMgodclaw | Details → | |
| 2 | SmolLM2-1.7B-Instruct1.7BHuggingFaceTB Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 338,72 tok/s TG Prefill 33.135 · TTFT 574 ms | 5× | 2x AMD Radeon RX 7900 XTXAMD Ryzen Threadripper PRO 3955WX 16-Cores · 63 GB RAM | vLLMgodclaw | Details → | |
| 3 | SmolLM2-1.7B-Instruct1.7BHuggingFaceTB Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 72,05 tok/s TG Prefill 13.702 · TTFT 223 ms | 1× | 2x AMD Radeon RX 7900 XTXAMD Ryzen Threadripper PRO 3955WX 16-Cores · 63 GB RAM | vLLMgodclaw | Details → | |
| 4 | SmolLM2-1.7B-Instruct1.7BHuggingFaceTB Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 3,10 tok/s TG Prefill 45 · TTFT 281.637 ms | 5× | Keine GPU (CPU-only) i4 x Cortex-A78C + 4 x Cortex-A55 · 15 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → | |
| 5 | SmolLM2-1.7B-Instruct1.7BHuggingFaceTB Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 2,89 tok/s TG Prefill 38 · TTFT 81.741 ms | 1× | Keine GPU (CPU-only) i4 x Cortex-A78C + 4 x Cortex-A55 · 15 GB RAM | llama.cppopenclaw_cliQ4_K_M | Details → |