How to get past limit of 77 tokens?

#77
by DDDano333 - opened

Token indices sequence length is longer than the specified maximum sequence length for this model (78 > 77). Running this sequence through the model will result in indexing errors

You could try to use the prompt batching described in this thread with the HF🧨 Diffusers version of SD3

the sd3 using there text_encoder, and I think it cannot use the same method as in the sdxl.

Sign up or log in to comment