Kolors
An open-source text-to-image generation model that supports LoRA adapter fine-tuning and is commonly used for diffusion model training and inference.
Timeline (last 90 days)
SpaceVision includes two LoRA adapters: the adapter for FLUX.2-klein (9B) is 38 MB in size, and the adapter for Kolors is 46.6 MB in size.
Kolors was developed by the Kuaishou team, is based on the U-Net architecture, uses ChatGLM3 as its text encoder, and supports Chinese prompts better than most English models.
All Facts (2)
SpaceVision includes two LoRA adapters: the adapter for FLUX.2-klein (9B) is 38 MB in size, and the adapter for Kolors is 46.6 MB in size.
50%UnverifiedKolors was developed by the Kuaishou team, is based on the U-Net architecture, uses ChatGLM3 as its text encoder, and supports Chinese prompts better than most English models.
50%