Partho Das

+1 running into the same error while trying to generate by calling model.generate() method in the getting started colab notebook.

found relevant issue huggingface/transformers#10160

this seems to be an issue with the `petals` library itself instead of the `transformers` library since replacing `AutoDistributedModelForCausalLM` with `AutoModelForCausalLM` seems to work fine

yes i agree! also gently pinging @borzunov here

Hi @NielsRogge, I would like to take on this. As Sylvain suggested, could you offer some context on how to go with this? Thanks :)

Hi, @tmabraham. If you're not currently working on this, I would be happy to take on it :)

The Latent Diffusion pipeline presently does not accept latents as an optional input; I'll open up a PR to add it.

made suggested changes. `vllm_support` is set to true by default and needs to be set false explicitly for unsupported models.

That would be an interesting thing to experiment with. Given a subject, generating stylistic clues as well as negative suggestions. The current [dataset](https://huggingface.co/datasets/poloclub/diffusiondb) we're using doesn't contain negative prompts; I'll...

Opened a Feature Request on A1111 repo. AUTOMATIC1111/stable-diffusion-webui/issues/5177