Zhuifeng_Official Model Motion Blur Fix_v1
UpdateOct 02, 2024 16:39PublishedOct 02, 2024 16:39
0
0
0
Zhuifeng_Official Model Motion Blur Fix_v1 - 1
Avatar
User_8x0jhm
Type
LoRA
Basic Model
FLUX.1 D
Published time
Oct 02, 2024 16:39
File Signature
ae6d5017b16a5f6aa115c63eef995d1fa7fdb195b6b25d799d697e1982593ef1
追风_官模运动模糊修复_v1.safetensors
128.29 MB

Trigger Words motion blur

image

A motion blur filter-style LoRA that has minimal impact on the base model, perfectly fixing the shortcomings of the official FLUX.1 model when generating motion blur style images.

When generating images in this style with the official model, the main subject is usually a blurry mess. With this LoRA added, it highlights the subject and keeps it crisp while enhancing the motion blur details, perfectly replicating the panning photography technique. image image Instructions:

  1. Base model: Use fp8-dev (including F.1 Compact _ CLIP and VAE included_Flux.1-Dev fp8) or fine-tuned base models based on fp8-dev. Types like alt, schnell, nf4, etc., are not recommended.

  2. Weight: 1 is sufficient.

  3. Trigger word: motion blur

  4. Image size: Any dimension works.

Gallery

No creation yet