Text-to-Image
Diffusers
Safetensors
QwenImagePipeline
nvidia
ModelOpt
AutoModel
qwen-image
dmd2
few-step
Instructions to use nvidia/Qwen-Image-Flash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Diffusers
How to use nvidia/Qwen-Image-Flash with Diffusers:
pip install -U diffusers transformers accelerate
import torch from diffusers import DiffusionPipeline # switch to "mps" for apple devices pipe = DiffusionPipeline.from_pretrained("nvidia/Qwen-Image-Flash", dtype=torch.bfloat16, device_map="cuda") prompt = "Astronaut in a jungle, cold color palette, muted colors, detailed, 8k" image = pipe(prompt).images[0] - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- Draw Things
- DiffusionBee
Commit History
-
Update OneIG-Bench-EN numbers to full five-dimension scores
Verified signatureThe previous OneIG-Bench-EN column reported the alignment dimension only (scored over 3 of 5 classes, 695 of 1120 prompts), which is not a OneIG score. Replaced with the unweighted mean of all five dimensions (alignment, text, reasoning, style, diversity) over the full 1120-prompt EN set, with a footnote noting OneIG publishes no official aggregate. Qwen-Image-Lightning is dashed as it was not re-scored on all five.eafac15
-
Format evaluation dataset fields
Verified signature8fba08a -
Finalize Qwen-Image-Flash model card
cbe1753
-
Restore Diffusers metadata for download tracking
Verified signaturebe31de3 -
Update Qwen-Image-Flash model card
Verified signature4873922
-
Add files using upload-large-folder tool
Verified signature20fe4b8 -
initial commit
Verified signaturec124532