OpenArt
OpenArt
Workflows

    My stuff

  • My Workflows

  • Liked Workflows

  • Following Workflows

Go to OpenArt main site
Upload workflow

Lumina2SDXL-L-Ollama (AutoPrompt) -IPA

5.0

1 reviews
8
2.4K
600
1
Description

This is a workflow that takes an uploaded image, automatically describes the image with Ollama (Llava 1.6 Works well) uses the Lumina ComfyUI wrapper and sends latent to SDXL lightning. Also uses IP Adapter to influence the final output. Can use up to 23gb VRAM but that can be fixed by getting rid of Ollama and just prompting normally. Also changing the SDXL-Lightning model to a model of choice, ie SD 1.5


https://github.com/kijai/ComfyUI-LuminaWrapper/tree/main


https://huggingface.co/Alpha-VLLM/Lumina-Next-SFT


https://ollama.com/library/llava


https://ollama.com/library/llava-phi3

Discussion

(No comments yet)

Loading...

Author

Indra's Mirror

Indra's Mirror

14
18.4K
226
82.6K

No reviews yet

  • - latest (a year ago)

  • - v20240619-200748

Primitive Nodes (5)

DownloadAndLoadGemmaModel (1)

DownloadAndLoadLuminaModel (1)

LuminaGemmaTextEncode (1)

LuminaT2ISampler (1)

Reroute (1)

Custom Nodes (23)

ComfyUI

  • - SaveImage (2)

  • - CheckpointLoaderSimple (1)

  • - PatchModelAddDownscale (1)

  • - CLIPTextEncode (2)

  • - VAEDecode (2)

  • - VAELoader (1)

  • - PreviewImage (2)

  • - EmptyLatentImage (1)

  • - LoadImage (1)

  • - KSampler (1)

ComfyUI Easy Use

  • - easy ipadapterApply (1)

ComfyUI Ollama

  • - OllamaVision (1)

ComfyUI_tinyterraNodes

  • - ttN textDebug (2)

mape's helpers

  • - mape Variable (2)

pythongosssss/ComfyUI-Custom-Scripts

  • - ShowText|pysssss (1)

WAS Node Suite

  • - Text Concatenate (1)

  • - Text Multiline (1)

Checkpoints (1)

MyModel\AmalgamationXL-V2-X-Tau-Lightning2.safetensors

LoRAs (0)