base_model:
- Wan-AI/Wan2.2-I2V-A14B
- Wan-AI/Wan2.2-T2V-A14B
tags:
- wan
- wan2.2
- accelerator
These are mixtures of WAN 2.2 and other WAN-like models and accelerators (with CLIP and VAE also included) to provide a fast, "all in one" solution for making videos as easily and quickly as possible. FP8 precision.
base: This is the first attempt and very "stable", but mostly WAN 2.1 with few WAN 2.2 features. sa_solver recommended.
V2: This is a more dynamic mixture with more WAN 2.2 features. sa_solver OR euler_a sampler recommended. Suffers from minor color shifts and noise in I2V, typically just at the start.
V3: This is a mixture of SkyReels and WAN 2.2, which should improve prompt adherence and quality. euler_a sampler recommended, beta scheduler. Suffers from minor color shifts and noise in I2V, typically just at the start.
V4: WAN 2.2 Lightning in the mix! euler_a/beta recommended. ClipVISION for I2V seems to be very weak due to being dropped in WAN 2.2, so your prompt may need to include more of what should stay in frame. Noise and color shifting generally fixed. WAN 2.1 LORA compatibility issues may be more present in this version.
You just need to use the basic ComfyUI "Load Checkpoint" node with these, as you can take the VAE, CLIP and Model all from one AIO safetensors. All models are intended to use 1 CFG and 4 steps. See sampler recommendations for each version.
v2+ models might be slightly less compatible with WAN 2.1 LORAs, as they bring in more WAN 2.2 features. Adjusting strengths of the LORAs may help (down if you see noise or artifacts). This merge will likely not be compatible with WAN 2.2 "high" LORAs, but should have good compatibility with WAN 2.1 + WAN 2.2 "low" LORAs.
Seems to work even on 8GB VRAM:
