Wan2.2 I2V A14B GGUF - A14B High
Recommended Prompts
a elf woman with long pointy ears black hair, Seductive, dark-red hair, cute, long hair, sweating, slim thighs, white hair, perfect round ass, silver pupils, is swimming inside a lagoon. She wears a bikini with dark-red trim and frills. The camera is partly under water and from behind. She is partly submerged. In the detailed background waterfall. She jiggles her ass and legs under water, girl in yoga pants and yoga bra, simulation realistic physics, with white cat ears and tail. her breasts are bouncing, small air bubbles go up around her legs., she is teasing, she is smiling. She moves her hips and breasts. sensual movements. She is flirting. Daytime
Recommended Parameters
samplers
steps
cfg
Tips
Use the high noise model as base and the low noise model as refiner when running Image To Video.
Set the Video Model to the high noise model and Video Swap Model to the low noise model with 0.5 Video Swap Percent in SwarmUI.
Reference CFG value is 3.5 for this model.
Use FPS of 16 by default, but increase to 24 if videos appear sped up or to look more normal.
Experiment with Sigma shift; default is 8, but some users recommend 5 for Image To Video.
Consider using Lightning-LoRAs to speed up processing (read instructions carefully).
Version Highlights
High version
Creator Sponsors
This GGUF file is a direct conversion of Wan-AI/Wan2.2-I2V-A14B and is available at https://huggingface.co/bullerwins/Wan2.2-I2V-A14B-GGUF/tree/main. Consider using it with Lightning-LoRAs for speed improvements.
Additional basic Workflow (ComfyUI) and Preset (SwarmUI) available here.
This GGUF file is a direct conversion of Wan-AI/Wan2.2-I2V-A14B / https://huggingface.co/bullerwins/Wan2.2-I2V-A14B-GGUF/tree/main
Consider using this with the Lightning-LoRAs for speed up (just read the description carefully!)
General usage (SwarmUI):
There's a 14B I2V (Image To Video), in a high+low noise pair
You're expected to run the high noise as a base and the low noise as a refiner
In the Image To Video params:
Set the regular Video Model to the high noise model,
and set the advanced Video Swap Model to the low noise model,
and leave Video Swap Percent at
0.5
Reference CFG range is
3.5This also supports the
Video End Frameinput to create a video that moves between two known places
For both 14B types:
FPS is
16, but loras or even parameter adjustments can change it to a more normal-looking24.Swarm will default to
24, but if your videos "feel sped up", change the FPS parameter to16.
Sigma shift may be worth experimenting with. The default is 8, but a wide range of values are functional.
Some users recommend
5for I2V
Added some basic Workflow (ComfyUI) and a Preset (SwarmUI): https://civitai.com/models/1823089?modelVersionId=2063163
Model Details
Model type
Base model
Model version
Model hash
Creator
Discussion
Please log in to leave a comment.
