Inference speed

[Kai-Sheng-caesar]

Hi, thanks for your great work!

I have a question regarding inference speed and whether my current performance is expected.

• GPU: NVIDIA A800 (80GB)
• Model: Qwen/Qwen-Image-Edit-2511
• LoRA: Qwen-Image-Edit-2511-Lightning-4steps-V1.0-fp32.safetensors
• Resolution: 640 × 480
• Batch size: 1

1. Is ~5s per 640×480 image on an A800 considered expected for this setup?
2. If this is expected, are there recommended ways to further speed up inference, ~0.5s per image?

Any guidance would be greatly appreciated. Thanks!