T5 v1.1 XXL encoder (the text_encoder_3 of Stable Diffusion 3.5 Medium) quantized to bitsandbytes NF4 (fp16 compute, wo kept in fp32). Load with T5EncoderModel.from_pretrained(repo, device_map={'': 'cuda:0'}) and pass it as text_encoder_3 to StableDiffusion3Pipeline. Weights are Google's T5 v1.1 XXL (Apache-2.0).

Downloads last month: -

Safetensors

Model size

5B params

Tensor type

F32

F16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for vftfnn/sd35-medium-t5-nf4

Base model

google/t5-v1_1-xxl

Quantized

(16)

this model