A workflow for Hunyuan video using a LLM prompt enhancer.
A short description can be given that would be enhanced by a LLM GGUG (using Searge_LLM, works for both Flux and Hunyuan)
After initial (lowres) video generation, you can enable the upscaler and interpolation for 1080p. Should work well with 12GB of VRAM.
Description
FAQ
Comments (2)
Confirmed working on 12gb of Vram, Great workflow. I love having LLM embedded into this, seems to provide much better results.
Can't make this work...I followed this: https://comfy.icu/extension/SeargeDP__ComfyUI_Searge_LLM and nope.... can't get it to work