🏆
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 | gemma-4-26B-A4B-it-assistant26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 74,51 tok/s TG Prefill 867 · TTFT 23.582 ms | 10× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 2 | gemma-4-26B-A4B-it-assistant26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 67,34 tok/s TG Prefill 795 · TTFT 12.430 ms | 5× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → | |
| 3 | gemma-4-26B-A4B-it-assistant26BGoogle Performance benchmarkTimebench 3 - Kombi (Prefill + Generation) | 32,97 tok/s TG Prefill 777 · TTFT 2.545 ms | 1× | AMD Radeon 8060S GraphicsAMD RYZEN AI MAX+ 395 w/ Radeon 8060S · 31 GB RAM | llama.cppgodclawQ4_K_M | Details → |