🏆
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 | Llama-3.2-1B-Instruct1BMeta Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 43,17 tok/s TG Prefill 281 · TTFT 8.805 ms | 1× | AMD Liverpool (PlayStation 4)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | Llama-3.2-1B-Instruct1BMeta Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 3,91 tok/s TG Prefill 193 · TTFT 12.786 ms | 1× | AMD Liverpool (PlayStation 4)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | Qwen3-0.6B0.6BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 3,80 tok/s TG Prefill 23 · TTFT 106.805 ms | 1× | AMD Liverpool (PlayStation 4)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 4 | Qwen3-4B4BQwen (Alibaba) Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 1,25 tok/s TG Prefill 4 · TTFT 474.264 ms | 1× | AMD Liverpool (PlayStation 4)AMD Jaguar 8-Core (PlayStation 4) · 5 GB RAM | llama.cppgodclawQ4_K_M | Details → |