Stable Diffusion 3.5 MEDIUM is BETTER than Large and Flux?
Added 2024-11-04 10:04:13 +0000 UTCIn this video, we take a comprehensive look into the newly released Stable Diffusion 3.5 Medium model. The video examines its features such as its uncensored nature, compatibility with consumer GPUs, and flexible licensing. We also compare its performance with Flux 1.1 Pro, Flux Dev, and Stable Diffusion 3.5 Large across various prompts, highlighting the strengths and weaknesses of each model in generating images. We discuss the model's prompting style, the reintroduction of negative prompts, and potential fine-tuning opportunities, all while addressing some encountered issues and potential settings adjustments needed for optimal results.
Links: Runpod: https://runpod.io?ref=49tc28ho
Runpod Comfy Template: https://runpod.io/console/deploy?template=p61ub3cnmv&ref=49tc28ho
Stable Diffusion 3.5 Large fp8 weights: https://huggingface.co/Comfy-Org/stable-diffusion-3.5-fp8/resolve/main/sd3.5_large_fp8_scaled.safetensors
SD 3.5 Medium: https://huggingface.co/stabilityai/stable-diffusion-3.5-medium/tree/main
SD 3.5 Large: https://huggingface.co/stabilityai/stable-diffusion-3.5-large/tree/main Sd 3.5 Large Turbo: https://huggingface.co/stabilityai/stable-diffusion-3.5-large-turbo
Workflows (FREE): https://huggingface.co/Comfy-Org/stable-diffusion-3.5-fp8/tree/main?ref=blog.comfy.org
Endangered AI Music Track: https://www.youtube.com/watch?v=Smk-lfVwpuk&t=1947s
