Qwen2.5-Coder-32B-Instruct
4 records across 2 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M5 Max | Qwen2.5-Coder-32B-Instruct | 4bit | oMLX | 27.8 | 772.3 | – | – | 1024 | 2026-10-18 | community | – | source |
| M5 Max | Qwen2.5-Coder-32B-Instruct | 4bit | oMLX | 26.3 | 798.3 | – | – | 4096 | 2026-10-18 | community | – | source |
| M4 Max | Qwen2.5-Coder-32B-Instruct | bf16 | oMLX | 7.2 | 216.4 | – | – | 4096 | 2026-10-10 | community | – | source |
| M4 Max | Qwen2.5-Coder-32B-Instruct | bf16 | oMLX | 6.7 | 170.2 | – | – | 16384 | 2026-10-10 | community | – | source |