Qwen3-4B-Instruct
2 records across 1 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M3 Pro | Qwen3-4B-Instruct | Q4_K_M | llama.cpp | 30.81 | – | – | – | – | 2026-09 | community | – | source |
| M3 Pro | Qwen3-4B-Instruct | Q8_0 | llama.cpp | 24.22 | – | – | – | – | 2026-09 | community | – | source |