Qwen3-VL-2B-Instruct
5 records across 2 hardware.
| hardware | model | quant | backend | tok/s | pp tok/s | W | ttft s | ctx/tg | date | class | flags | source |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| M5 Pro | Qwen3-VL-2B-Instruct | 4bit | oMLX | 189.1 | 4599.0 | – | – | 1024 | 2026-10-08 | community | – | source |
| M4 | Qwen3-VL-2B-Instruct | 4bit | oMLX | 88.3 | 1120.0 | – | – | 1024 | 2026-10-10 | community | – | source |
| M4 | Qwen3-VL-2B-Instruct | 4bit | oMLX | 66.3 | 1016.0 | – | – | 4096 | 2026-10-10 | community | – | source |
| M4 | Qwen3-VL-2B-Instruct | 4bit | oMLX | 50.5 | 895.3 | – | – | 8192 | 2026-10-10 | community | – | source |
| M4 | Qwen3-VL-2B-Instruct | 4bit | oMLX | 34.4 | 720.3 | – | – | 16384 | 2026-10-10 | community | – | source |