Downloads · 30 days
307
6% of all-time downloads
PJMixers-Images/lightx2v_Qwen-Image-Lightning-4step-8step-Merge
lightx2v_Qwen-Image-Lightning-4step-8step-Merge is a text-to-image model from PJMixers-Images. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as apache-2.0.
Downloads · 30 days
307
6% of all-time downloads
All-time downloads
5.3K
Public
Repo size
2.6 GB
Likes
13
Public
Click a slice to open those files.
.safetensors2.5 GB · 99%
From the Hugging Face model README
The 4-step LoRA does fairly well at 4 steps, but it cannot go higher than 4 steps without overcooking the image, and even at 4 steps the image feels a little cooked.
<p align="center"> <img src="examples/comparisons/Comparison_00001_.png" height="480px"/> <span style="font-size: small;">4-step lora | 4 steps vs. 8 steps vs. 16 steps | [1536x1536, no shift, lora strength 1, deis, kl_optimal, seed 187]</span> </p>The 8-step LoRA on the other hand is very undercooked at 4 steps, still a little undercooked at 8 steps, but handles higher step counts like 16 really well, but feels a little undercooked overall.
<p align="center"> <img src="examples/comparisons/Comparison_00002_.png" height="480px"/> <span style="font-size: small;">8-step lora | 4 steps vs. 8 steps vs. 16 steps | [1536x1536, no shift, lora strength 1, deis, kl_optimal, seed 187]</span> </p>Merging these two together results in being able to do 16 without overcooking or undercooking. It feels just about right, especially if you load at 90% strength.
<p align="center"> <img src="examples/comparisons/Comparison_00005_.png" height="480px"/> <span style="font-size: small;">merged lora | 4 steps vs. 8 steps vs. 16 steps | [1536x1536, no shift, lora strength 1, deis, kl_optimal, seed 187]</span> </p>