---

name: curated_wan_2_2_i2i_14b_gguf_4_step_lora_upscale description: > Curated ComfyUI workflow imported from Wan 2.2 I2I 14b GGUF + 4 Step Lora + Upscale.json. homepage: https://github.com/steliosot/comfy-agent metadata.clawdbot.os: ["darwin", "linux"] metadata.clawdbot.requires.bins: ["python3"] metadata.clawdbot.requires.env: ["COMFY_URL", "COMFY_AUTH_HEADER"] metadata.clawdbot.files: ["workflow.json", "skill.py", "skill.yaml", "scripts/*"] metadata.clawdbot.tags: ["comfyui", "workflow", "curated", "video_t2v_i2v_avatar"] metadata.clawdbot.category: "media-generation" metadata.clawdbot.input_type: "image/png" metadata.clawdbot.output_type: "image/png" metadata.clawdbot.output_can_feed_into: [] metadata.clawdbot.accepts_input_from: [] metadata.clawdbot.priority: 70

curated_wan_2_2_i2i_14b_gguf_4_step_lora_upscale

Curated workflow skill generated from Wan 2.2 I2I 14b GGUF + 4 Step Lora + Upscale.json.

Capability Family

• video_t2v_i2v_avatar

Inputs

• Optional runtime overrides supported by run(...):
  • prompt
  • negative_prompt
  • width, height
  • seed, steps, cfg
  • sampler_name, scheduler, denoise
  • server, headers, api_prefix

Outputs

• Returns JSON with:
  • status
  • prompt_id
  • output_images (includes image/video entries reported by Comfy history)

Model Requirements

• diffusion_model: Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf -> models/diffusion_models
• vae: wan_2.1_vae.safetensors -> models/vae
• clip: umt5_xxl_fp8_e4m3fn_scaled.safetensors -> models/clip
• upscale_model: 4x_NMKD-Siax_200k.pth -> models/upscale_models

Custom Node Requirements

• ComfyUI-GGUF
• comfyui-easy-use
• comfyui-kjnodes
• rgthree-comfy

Links Extracted From Workflow Notes

• https://discord.com/invite/gggpkVgBf3
• https://huggingface.co/Akumetsu971/SD_Anime_Futuristic_Armor/resolve/main/4x_NMKD-Siax_200k.pth?download=true
• https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/text_encoders/umt5_xxl_fp8_e4m3fn_scaled.safetensors?download=true
• https://huggingface.co/Comfy-Org/Wan_2.1_ComfyUI_repackaged/resolve/main/split_files/vae/wan_2.1_vae.safetensors?download=true
• https://huggingface.co/QuantStack/Wan2.2-T2V-A14B-GGUF/resolve/main/LowNoise/Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf?download=true
• https://huggingface.co/QuantStack/Wan2.2-T2V-A14B-GGUF/tree/main/LowNoise
• https://huggingface.co/lightx2v/Wan2.2-Lightning/resolve/main/Wan2.2-T2V-A14B-4steps-lora-rank64-Seko-V1.1/low_noise_model.safetensors?download=true
• https://huggingface.co/lightx2v/Wan2.2-Lightning/tree/main/Wan2.2-T2V-A14B-4steps-lora-rank64-Seko-V1.1
• https://www.youtube.com/@pixaroma
• https://www.youtube.com/watch?v=CgLL5aoEX-s

Source

• Original: comfy-data/workflows/Wan 2.2 I2I 14b GGUF + 4 Step Lora + Upscale.json

Routing Metadata

• Family: video_t2v_i2v_avatar
• Input modalities: image, text_prompt
• Output modalities: image/png
• Model families: sd3, wan
• Node count: 32
• Complexity score: 9
• Resource profile: very_high
• Estimated runtime: heavy (can exceed 6 min; best on high-VRAM GPU)
• Max latent resolution hint: NonexNone
• Max sampler steps hint: 8

Detected Models

• diffusion_model: Wan2.2-T2V-A14B-LowNoise-Q8_0.gguf -> models/diffusion_models
• vae: wan_2.1_vae.safetensors -> models/vae
• clip: umt5_xxl_fp8_e4m3fn_scaled.safetensors -> models/clip
• upscale_model: 4x_NMKD-Siax_200k.pth -> models/upscale_models

Detected Custom Nodes

• ComfyUI-GGUF
• comfyui-easy-use
• comfyui-kjnodes
• rgthree-comfy

Runtime Warnings

• Large model(s) detected; ensure enough VRAM and disk space.
• Multiple custom nodes required; verify node installation/version compatibility.
• Video workflow: usually slower and VRAM-intensive than still-image workflows.