Spaces:

JitRoy2024
/

Qwen_Img_Space

Running on Zero

110 kB

Ctrl+K

5 contributors

Jit2024

Fix ZeroGPU RuntimeError timeout and 89% hang by adding FP8 quantization. Reduce VRAM footprint to <29GB to prevent OOM on standard 48GB L40S nodes, and restore root .to(cuda) for rapid initialization.

2ec5462 about 1 month ago

qwenimage
Derive output aspect ratio from first image, not last 3 months ago
.gitattributes

1.67 kB
Upload 3 files 10 months ago
.gitignore

559 Bytes
Add .claude/ to gitignore 6 months ago
AGENT.md

6.64 kB
Agent guidelines 9 months ago
CLAUDE.md

6.08 kB
Upgrade to Qwen-Image-Edit-2511 with Phr00t v18 transformer 8 months ago
README.md

710 Bytes
Add debug diagnostic dashboard 3 months ago
app.py

16 kB
Fix ZeroGPU RuntimeError timeout and 89% hang by adding FP8 quantization. Reduce VRAM footprint to <29GB to prevent OOM on standard 48GB L40S nodes, and restore root .to(cuda) for rapid initialization. about 1 month ago
optimization.py

2.5 kB
Bisect AOTI compile duration upward: 600 → 1000 3 months ago
requirements.txt

157 Bytes
Pin kernels<0.15.1 to dodge required revision/version arg break about 2 months ago