Frank Frazetta style lora. Hidream was clearly already pretrained on his images. This only took 800 steps to get it to produce images in his style.
Qwen 2.1 - This is a LoKr not a LORA. It should import the same way in comfyui. CFG should be between 2-3.5. 3 and 3.5 work well.
Description
FAQ
Comments (3)
would you mind posting the dataset? I would like to train a lora with this for Wan2.1 and Wan2.2.
Details
Downloads
719
Platform
CivitAI
Platform Status
Available
Created
8/12/2025
Updated
10/28/2026
Deleted
-
Trigger Words:
Frank Frazetta style.
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.