Wan 2.2 First Last Frame — v1.0 TensorRT
My latest workflow for First Last Frame with Wan 2.2.
Runtime profile
- Access
- External workflow
- Status
- Listed
- Version
- Gen3dxFLFWan2.2.json
- Source
- Civitai
- Updated
- Nov 3, 2026
Source description
No release description captured.
What you need
- Model files
- 38.2 GiB known subtotal
Remaining hardware or download-size estimates are not available yet.
Requirements
Models
6Text encoder · 6.27 GB · SAFETENSORS · U8 + FP16 + FP32 + FP8 (tensor header) · Hugging Face · Comfy-Org/Wan_2.1_ComfyUI_repackaged
Unresolved model selection: LoadUpscalerTensorrtModel
Not resolvedUnknown
VAE · 242 MB · SAFETENSORS · BF16 (tensor header) · Hugging Face · Comfy-Org/Wan_2.1_ComfyUI_repackaged
Diffusion model · 30.5 GB · SAFETENSORS · Hugging Face · Comfy-Org/Wan_2.1_ComfyUI_repackaged
LoRA · 585 MB · SAFETENSORS · FP16 (unverified) · Unknown
LoRA · 585 MB · SAFETENSORS · FP16 (unverified) · Unknown
Node packs
7Node pack · GitHub · pythongosssss/ComfyUI-Custom-Scripts
- Found
Node pack · GitHub · Smirnov75/ComfyUI-mxToolkit
ComfyUI-Rife-Tensorrt
FoundNode pack · Registry
ComfyUI-Upscaler-Tensorrt
FoundNode pack · Registry
comfyui-videohelpersuite
FoundNode pack · Registry
rgthree-comfy
FoundNode pack · Registry
- Found
Node pack · GitHub · orssorbit/ComfyUI-wanBlockswap
AI-generated commentary
AI-generated explanation based on source and configuration details. Suggestions are clearly labeled.
Purpose
Use Wan 2.2 First Last Frame to generate video guided by first and last frame images; it supports LoRA loading, block swapping, interpolation, and upscaling, with Wan Video 2.2 I2V-A14B as its base model.
Flow
The workflow combines first-and-last-frame guidance with LoRA loading, block swapping, interpolation, and upscaling.
Two versions are available: one with TensorRT interpolation and upscaling, and one with standard interpolation and upscaling.
Inputs and outputs
Use first and last frame images to guide and control the scene and action during video generation.
Interpolation and upscaling are included as finishing touches for video generation.
Setup
Choose the TensorRT version with an RTX50xx series card; use the standard version with other video cards.
Suggestion · not verified
Dependencies
The workflow names these model files: umt5_xxl_fp8_e4m3fn_scaled.safetensors, wan_2.1_vae.safetensors, wan2.1_flf2v_720p_14B_fp16.safetensors, wan2.2_i2v_high_noise_14B_fp8_scaled.safetensors, and wan2.2_i2v_low_noise_14B_fp8_scaled.safetensors.
The workflow names these node packs: comfyui-custom-scripts, comfyui-mxtoolkit, ComfyUI-Rife-Tensorrt, ComfyUI-Upscaler-Tensorrt, comfyui-videohelpersuite, rgthree-comfy, and wanblockswap.
Configuration
Block swapping is included to help with VRAM.
Cautions
Before use, verify that wanblockswap is available in your setup.
Suggestion · not verified
Before use, verify that your setup has the exact high-noise and low-noise model files named above.
Suggestion · not verified
Does this need to be edited?
Sign in to send an edit request.
Sources
1 source
Sources
1 sourceDownload files
Model artifacts
2 artifacts
Model artifacts
2 artifactsSource excerpts
3 excerpts
Source excerpts
3 excerptsSource context: 704 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B
Source context: 672 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B
Source context: 580 downloads · Type Workflows · Base model Wan Video 2.2 I2V-A14B
