InvokeAI: [m1] [ti] textual inversion embedding loaded but * doesn't work

GPU: M1 CPU arch: arm OS: MacOS Python: miniconda Branch: development / release-candidate-2 Commit: [b296933] / [3d7bc07]

ugly-sonic.bin loaded: >> Model loaded in 10.24s Added terms: *, <ugly-sonic> >> Setting Sampler to k_lms

then: dream> photo of <ugly-sonic> produces: RuntimeError: Tensors must have same number of dimensions: got 2 and 1 >> Could not generate image.

OR: dream> photo of * produces: photo of random guy.

the same for other models or styles from hugging face. bin size is correct.

About this issue

  • Original URL
  • State: closed
  • Created 2 years ago
  • Comments: 19 (14 by maintainers)

Commits related to this issue

Most upvoted comments

Seems to work? Same seed:

<cat-toy> 000333 3969801327

vs cat-toy 000334 3969801327

Thank you guys, testing…

@Any-Winter-4079 this will be more obvious: https://huggingface.co/sd-concepts-library/cat-toy

your zip is ok, but not the output.