Home Creators Posts Import Register Favorites Logout

Content

https://youtu.be/RKxrXkVpPoE

https://github.com/lllyasviel/Omost

Want an LLM to generate prompts for you? Maybe it'll get it Omost right! Yes, the puns abound with this latest release from lllyasviel. With more of a large language model style chat interface, the LLM first generates special canvas code before handing over to SDXL for the image generation.

This means you can easily put different things in different areas of your image, like in the example below:

So long at you've got at least 8 GB VRAM, you should be golden. It does say you need an Nvidia GPU, but who knows... maybe ROCm pytorch works too?

If you have >18GB VRAM I suggest enabling high VRAM mode in lib_omost/memory_management.py - though I also had to disable both of the "unloads" in gradio_app.py as well. I'm sure it will change in the future though, as it's only been out for a couple of days!

It certainly seems to be a lot of fun, mostly because it just works with whatever you give it!

Files

Comments

Lord Banzai

Also, Check "ComfyUI-IF_AI_tools"

Serious S.H.I.T. (edited)

Comment edits

2026-09-14 18:00 im getting this - J:\ai\Omost\lib_omost\pipeline.py:64: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor). alphas_cumprod = torch.tensor(np.cumprod(alphas, axis=0), dtype=torch.float32) Unload to CPU: CLIPTextModel Unload to CPU: UNet2DConditionModel Unload to CPU: AutoencoderKL Unload to CPU: CLIPTextModel Unused kwargs: ['_load_in_4bit', '_load_in_8bit', 'quant_method']. These kwargs are not used in . Loading checkpoint shards: 100%|███████████████████████████████████████████████████████████████████| 2/2 [00:05<00:00, 2.97s/it] Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. You shouldn't move a model that is dispatched using accelerate hooks. no idea what it means , im using an rtx3060 12gb - any idea how to fix it?
2024-06-02 11:00

NerdyRodent

And then after that I assume it says: Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. Running on local URL: http://0.0.0.0:7860