This workflow uses SVI 2 PRO to extend existing videos: just load your video, add a prompt, and watch it seamlessly generate the continuation.
Runtime profile
Source description
SVI2Pro+FLFVideoExtension
Allows you to extend existing videos using keyframes.
SVI2PRO FLF V1.1
Added a switch to enable/disable end_samples (Last Frame) directly, without manually connecting/disconnecting inputs. This toggles the First/Last Frame functionality: if disabled, the node reverts to the base SVI 2 PRO behavior. (Requires updating the Wan-SVI2Pro-FLF nodes to the latest version)
Added contrast adjustment widget, useful if your output looks washed out.
SVI2PRO FLF V1
Workflow for ComfyUI that combines SVI 2 Pro motion continuity with Wan 2.2 First/Last Frame (FLF) style control over the end of a clip, enabling smooth video generation from a sequence of frames.
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Use this ComfyUI workflow to extend an existing video: provide a video and a prompt, and it can generate a continuation with First/Last Frame keyframe control.
First/Last Frame control combines SVI 2 PRO motion continuity with Wan 2.2 control over the end of a clip.
The experimental 3-Part Loop Mode moves from Start to Loop entry, repeats the loop action, then moves from Loop exit to the Final frame.
Video-extension mode takes an existing video and a prompt, then outputs a continuation.
3-Part Loop Mode takes Start, Loop, and End images plus three prompts and automates a sequence ending at the Final frame.
The listed base model is Wan Video 2.2 I2V-A14B, and the workflow version is SVI2PRO VideoExtension.
Before use, check that these ComfyUI node packs are available: comfyui-frame-interpolation, comfyui-gguf, comfyui_essentials, comfyui-easy-use, comfyui-custom-scripts, comfyui-kjnodes, comfyui-videohelpersuite, comfyui-customnodes-template, comfyui_layerstyle, rgthree-comfy, and
Suggestion · not verified
Also check for these model files: umt5_xxl_fp8_e4m3fn_scaled.safetensors, wan22EnhancedNSFWCameraPrompt_nsfwFASTMOVEV2Q8L.gguf, wan2.2_i2v_lightx2v_4steps_lora_v1_high_noise.safetensors, SVI_v2_PRO_Wan2.2-I2V-A14B_HIGH_lora_rank_128_fp16.safetensors, and wan_2.1_vae.safetensors.
Suggestion · not verified
The list also includes RealESRGAN_x4plus.safetensors, wan2.2_i2v_lightx2v_4steps_lora_v1_low_noise.safetensors, wan22EnhancedNSFWCameraPrompt_nsfwFASTMOVEV2Q8H.gguf, rife49.pth, and SVI_v2_PRO_Wan2.2-I2V-A14B_LOW_lora_rank_128_fp16.safetensors.
Suggestion · not verified
The First/Last Frame switch can enable or disable the Last Frame input; when disabled, the node reverts to base SVI 2 PRO behavior.
A contrast-adjustment widget is available for cases where the output looks washed out.
In 3-Part Loop Mode, enabling increment can vary the repeating loop action per cycle; match each prompt to its corresponding image transition.
Suggestion · not verified
Before setup, verify which packages to install for comfyui-gguf and ComfyMath, and verify the sources for wan22EnhancedNSFWCameraPrompt_nsfwFASTMOVEV2Q8L.gguf, wan22EnhancedNSFWCameraPrompt_nsfwFASTMOVEV2Q8H.gguf, and SVI_v2_PRO_Wan2.2-I2V-A14B_HIGH_lora_rank_128_fp16.safetensors
Suggestion · not verified
Before using First/Last Frame control, verify that the Wan-SVI2Pro-FLF nodes are up to date. The 3-Part Loop Mode is described as experimental, and no license is listed.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 sourceSource excerpts
2 excerptsSource context: 1794 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B
SVI2PRO VideoExtension
This workflow uses SVI 2 PRO to extend existing videos: just load your video, add a prompt, and watch it seamlessly generate the continuation.
3-Stage Loop Automation
Experimental: 3-Part Loop Mode
Uses 3 images (Start / Loop / End) and 3 prompts to automate sequences:
Start → Loop entry
Repeating loop action (varies per cycle if increment is enabled)
Loop exit → Final frame
Ensure prompts match their corresponding image transitions.
If you like these workflows, I’d love to see your results, please share your videos in the Gallery or link them in the comments.
Repository:
Recommended models for these workflows:
SFW:
NSFW:
Estimated VRAM requirement
Estimate unavailable
8.86 GB across 5 of 10 model files. Model file total + 25% loading overhead + 2 GB execution buffer, rounded up.
Requirements
21 requirementsUpscaler · 63.8 MB · SAFETENSORS · Hugging Face · Comfy-Org/Real-ESRGAN_repackaged
Checkpoint · Hugging Face · VMTamashii/rife49
LoRA · Hugging Face
LoRA · Hugging Face · alphaghost13/SVI_v2_PRO_Wan2.2-I2V-A14B_LOW_lora_rank_128_fp16
Text encoder · 6.27 GB · Hugging Face · Comfy-Org/Wan_2.1_ComfyUI_repackaged
VAE · 242 MB · Hugging Face · Comfy-Org/Wan_2.2_ComfyUI_Repackaged
LoRA · 1.14 GB · Hugging Face · Comfy-Org/Wan_2.2_ComfyUI_Repackaged
LoRA · 1.14 GB · SAFETENSORS · Hugging Face · Comfy-Org/Wan_2.2_ComfyUI_Repackaged
wan22EnhancedNSFWCameraPrompt_nsfwFASTMOVEV2Q8H.gguf
Not resolvedCheckpoint · Unknown
wan22EnhancedNSFWCameraPrompt_nsfwFASTMOVEV2Q8L.gguf
Not resolvedCheckpoint · Unknown
ComfyMath
PossibleNode pack · Registry
Node pack · Registry
Source context: 1709 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B
Node pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry
comfyui-gguf
PossibleNode pack · Registry
Node pack · Registry
Node pack · Registry
Node pack · Registry