Vision-language · Alibaba Qwen
Qwen3 VL 30B-A3B
MoE vision-language model, 3B active; good fit for 24–32 GB.
Datasheet
Maker | Alibaba Qwen |
|---|---|
Parameters | 31.07 BVerifiedsafetensors.total from the HF API Source |
Licence | Apache-2.0Verified |
Access | Public download |
Task (HF pipeline) | image-text-to-text |
First published on HF | 2025-09-30 |
Last updated on HF | 2025-11-26 |
Weights in memory | 62 GB at 16-bit · 19 GB at ~4.85 bitsEstimateParameters × bits ÷ 8. Excludes context cache and runtime overhead; not every model has a 4-bit build. |
Benchmark
Not yet measured
MMMU (val)
No comparable public score for this benchmark yet (Scores published as images only). We will not estimate one.
Used in
The description summarises the model card at https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct. Parameters and licence are read from the Hugging Face API; the weights-in-memory line is our arithmetic, not a measurement.