Wan 2.1 FusionX T2V GGUF workflow (Fusionx 워크플로우 공유- AI는 에이원)

This workflow is for text-to-video generation with a Wan2.1 T2V GGUF setup: it is described as using text, a LoRA, and a GGUF model to generate video.

The description names the GGUF model as QuantStack/Wan2.1_T2V_14B_FusionX-GGUF and the LoRA file as Wan21_T2V_14B_lightx2v_cfg_step_distill_lora_rank32.safetensors.

Its stated task is text-to-video generation, with video as the output.

The workflow is published by A01demort on Civitai.

The listed model files are umt5_xxl_fp8_e4m3fn_scaled.safetensors, umt5_xxl_fp16.safetensors, wan_2.1_vae.safetensors, and Wan2.1-14B-T2V-FusionX-Q5_K_M.gguf.

The listed node packs are comfyui_layerstyle, comfyui-easy-use, comfyui-kjnodes, comfyui-multigpu, comfyui-videohelpersuite, and rgthree-comfy.

The declared base model is Wan Video 14B t2v.

To vary the setup, consider swapping the LoRA and GGUF model named in the description.

Sugestão · não verificado

Wan2.1-14B-T2V-FusionX-Q5_K_M.gguf has no provider source specified, so verify that you have the file before use.

Sugestão · não verificado

No license is listed for the workflow, so verify the applicable usage terms before use.

Sugestão · não verificado