Frame-by-Frame Image-to-Image Stable Diffusion workflow — v2.0
Added Depth and OpenPose controlnets and AnimateDiff
Runtime profile
- Access
- External workflow
- Status
- Listed
- Version
- Frame-by-Frame img2img (animatediff & controlnet).json
- Source
- Civitai
- Updated
- Nov 3, 2026
What you need
- Model files
- 1.9 GiB known subtotal
Remaining hardware or download-size estimates are not available yet.
Requirements
Models
12control_v11f1p_sd15_depth.pth
Not resolvedControlNet · Unknown
ControlNet · 1.35 GB · PTH · Unknown
- File specified
LoRA · 218 MB · Civitai · 245590
dw-ll_ucoco_384.onnx
Not resolvedCheckpoint · Unknown
hsxl_temporal_layers.safetensors
Not resolvedCheckpoint · Unknown
jucika-000030.safetensors
Not resolvedLoRA · Unknown
Vision model · Hugging Face · MiaoshouAI/Florence-2-large-PromptGen-v2.0
VAE · 319 MB · SAFETENSORS · Unknown
Takanashi Kiara (illust).safetensors
Not resolvedLoRA · Unknown
Tavern.safetensors
Not resolvedLoRA · Unknown
waiNSFWIllustrious_v150.safetensors
Not resolvedCheckpoint · Unknown
yolox_l.onnx
Not resolvedCheckpoint · Unknown
Node packs
9comfyui_controlnet_aux
PossibleNode pack · Registry
comfyui-advanced-controlnet
FoundNode pack · Registry
comfyui-animatediff-evolved
FoundNode pack · Registry
Node pack · GitHub · kijai/ComfyUI-Florence2
comfyui-image-saver
FoundNode pack · Registry
comfyui-logic
FoundNode pack · Registry
comfyui-videohelpersuite
FoundNode pack · Registry
RES4LYF
FoundNode pack · Registry
rgthree-comfy
FoundNode pack · Registry
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Purpose
Use this workflow to process a video frame by frame with img2img, LoRAs, and optional autocaptioning, then recombine the processed frames into a new video.
Flow
Each frame is processed separately before the processed frames are recombined into a new video.
Inputs and outputs
Input: a video. Output: a new video made from the processed frames.
Frame processing uses img2img with LoRAs and can include autocaptioning.
Dependencies
Named node packs include comfyui-advanced-controlnet, RES4LYF, rgthree-comfy, comfyui-logic, and comfyui-image-saver.
It also names comfyui-videohelpersuite, comfyui-animatediff-evolved, comfyui-florence2, and comfyui_controlnet_aux.
Named model and file names include control_v11f1p_sd15_depth.pth, control_v11p_sd15_openpose.pth, hsxl_temporal_layers.safetensors, MiaoshouAI/Florence-2-large-PromptGen-v2.0, SD/pppanimixVAE_il.safetensors, and waiNSFWIllustrious_v150.safetensors.
Configuration
The listed base model is Illustrious.
Cautions
Before using it, verify which installed package provides comfyui-florence2 and comfyui_controlnet_aux.
Suggestion · not verified
Before using it, verify that these files are available: control_v11f1p_sd15_depth.pth, control_v11p_sd15_openpose.pth, hsxl_temporal_layers.safetensors, SD/pppanimixVAE_il.safetensors, and waiNSFWIllustrious_v150.safetensors.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 source
Sources
1 sourceWorkflow instructions
Author notes from Frame-by-Frame img2img (animatediff & controlnet).json · selected release
⚠️ BATCH SIZES OVER 1 DO NOT WORK ⚠️ Disable by bypassing (right click and then left click the arrow going around the circle) the two purple nodes "Florence Model" and "Florence2Run"
Download files
Model artifacts
2 artifacts
Model artifacts
2 artifactsFrame-by-Frame img2img (animatediff & controlnet).json
json · 56.7 KB · SHA-256 3651dfe26892…ce9f
Download fileSource excerpts
3 excerpts
Source excerpts
3 excerptsSource context: 346 downloads · Type Workflows · Base model Illustrious
Source context: 329 downloads · Type Workflows · Base model Illustrious
Source context: 290 downloads · Type Workflows · Base model Illustrious
