Wan
Collection
4 items • Updated
Wan2.2-TI2V-5B-FP8
FP8 safetensors artifact of Wan2.2-TI2V-5B.
MECHA Forge
MECHA — Model Engineering & Contextual Hearth Assembly
This package was created in the MECHA Forge.
Quantization Details
- Base model:
Wan2.2-TI2V-5B - Format: FP8
- Quant backend: TorchAO
- TorchAO method:
float8_weight_only
Load with:
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained(
"yourname/Wan2.2-TI2V-5B-FP8",
device_map="auto",
)
Requires a compatible torchao installation.
- Downloads last month
- 37
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support