kijai/ComfyUI-WanVideoWrapper

ComfyUI wrapper nodes for WanVideo is a set of custom nodes that integrate the WanVideo video generation model into the ComfyUI interface.

View on GitHub ↗Jump to charts ↓

Summary Information

Updated 10 minutes ago
Added to GitGenius on September 9th, 2026
Created on February 25th, 2025
Open Issues & Pull Requests: 1,282 (+0)
GitHub issues: Enabled
Number of forks: 693
Total Stargazers: 6,694 (+0)
Total Subscribers: 74 (+0)

Repository Insights (GitGenius)

Most active contributors

Sign in to see contributor activity.

Related repositories by overlapping contributors

No overlapping-contributor repos identified yet.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 1,239
New in 7 days: 1
Closed in 7 days: 0
Avg open age: 401 days
Stale 30+ days: 1,235
Stale 90+ days: 1,211

Recent activity

Opened in 7 days: 1
Closed in 7 days: 0
Comments in 7 days: 0
Events in 7 days: 0

Top labels

No label distribution available yet.

Detailed Description

ComfyUI wrapper nodes for WanVideo is a set of custom nodes that integrate the WanVideo video generation model into the ComfyUI interface.

The tool addresses the challenge of using WanVideo within ComfyUI's node-based workflow system. It wraps WanVideo and related models to make them accessible as ComfyUI nodes, allowing users to incorporate video generation into their ComfyUI graphs. The wrapper includes optimizations focused on memory efficiency, particularly around VRAM usage during model inference and LoRA weight handling.

Developers should adopt this tool only if they need features or model versions not yet available through WanVideo's native ComfyUI support. The wrapper is most useful for those already working within ComfyUI who want to experiment with WanVideo models before native integration becomes available. The project explicitly acknowledges that native implementations should be preferred when available, positioning itself as a bridge solution rather than a permanent alternative.

The project is marked as perpetually work in progress. Development activity shows active attention to memory optimization challenges, with recent updates addressing VRAM efficiency through changes to how LoRA weights are handled and how torch.compile interacts with the model. The maintainer has documented specific workarounds for Windows-related Triton cache issues that can cause excessive memory use on first runs after updates. The project has implemented moderation on issue reporting due to spam, indicating active maintenance but also suggesting the need to manage community engagement carefully.