Qwen3.8-40B-Ultimate
- Task
- Vision-language (image + text to text)
- Category
- Image
- License
- Apache 2.0
- Commercial use
- Yes
Permissive license: commercial use is allowed; keep the copyright and license notice. - Size
- About 40 billion parameters
- Memory to load
- ~96 GB (estimate)
- Format
- FP16
- Runs on
- GPU
- Access
- Gated: accept the publisher's terms on Hugging Face
- Downloads (last 30 days)
- 0
- Likes
- 0
- Last updated
- October 22, 2026
How to run it
These are full-precision weights, usually loaded with Hugging Face Transformers or Diffusers on a GPU.
Plan for roughly 96 GB of memory to load it (our estimate from its size and format).
Downloads are gated: accept the publisher's terms on Hugging Face first.
The model card on Hugging Face has the exact usage code.
Where to run it
In the cloud
Needs about 96 GB: more than one GPU, for example 2 × 80 GB A100 or H100.
- RunPod · rent a GPU by the hour (affiliate link)
Related vision-language (image + text to text) models
Data from Hugging Face's public API, last checked October 22, 2026. Downloads are Hugging Face's count for the last 30 days.