Wan2.2 models
7 Wan2.2 models on Hugging Face, from 5.0B to 16.3B parameters. At the precision each one is published in, the smallest needs about 11.2 GB of VRAM (LongLive-2.0-5B-Diffusers, cheapest live fit: RTX 5060 Ti) and the largest about 63.9 GB (Wan2.2-I2V-A14B-Diffusers, cheapest live fit: A100). The cheapest way to run LongLive-2.0-5B-Diffusers is $0.110/hr.
Wan2.2 models
| Model | Parameters | Published as | VRAM needed | Live GPU fit | Est. $/hr |
|---|---|---|---|---|---|
| LongLive-2.0-5B-Diffusers | 5.0B | BF16 | 11.2 GB | RTX 5060 Ti | $0.110/hr |
| Wan2.2-TI2V-5B-Diffusers | 5.0B | F32 | 22.4 GB | RTX A5000 | $0.176/hr |
| Wan2.2-T2V-A14B-Diffusers | 14.3B | F32 | 63.9 GB | A100 | $1.21/hr |
| Wan2.2-I2V-A14B-Lightning-Diffusers | 14.3B | BF16 | 31.9 GB | RTX 4080 Super | $0.338/hr |
| Wan2.2-I2V-A14B-Diffusers | 14.3B | F32 | 63.9 GB | A100 | $1.21/hr |
| Wan2.2-S2V-14B-Diffusers | 16.3B | BF16 | 36.4 GB | RTX A6000 | $0.363/hr |
| Wan2.2-S2V-14B | 16.3B | BF16 | 36.4 GB | RTX A6000 | $0.363/hr |
VRAM is for the precision the model is published in, with the same overhead and an 8,192-token context assumed on every page; see the methodology. The fit shown is the lowest-priced single GPU type that holds the model at that precision, or the lowest-priced multi-GPU set (up to 8) when none does.
More from Wan-AI
- Wan2.1 (11 models)
