A - Lets you create an image
B1 - You can Drop an image in here to do Image to Text - Just install Ollama on your machine and it runs
B2 - Image to Text with Florence 2 - Really good and accurate
C - Ollama Text Generation - Need to be inspired? Get LLM to create you next prompt'
D - Supir Upscale - Upscale your image for mor details
E - Image Upscale with Ultimate Upscaler
Files you need to run the workflow and where to download them from:
FLUX dev FP16 (23.8 GB) : https://huggingface.co/OwlMaster/realgg/resolve/main/flux1-dev.safetensors
FLUX dev FP8 (11.9 GB) - If you have 16GB VRAM or Less use this
https://huggingface.co/Kijai/flux-fp8/resolve/main/flux1-dev-fp8.safetensors
FLUX schnell FP8 (11.9 GB):
https://huggingface.co/Kijai/flux-fp8/blob/main/flux1-schnell-fp8.safetensors
Download Clip L (250 MB) : https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/clip_l.safetensors
T5 XXL FP16 (9.8 GB) : https://huggingface.co/comfyanonymous/flux_text_encoders/resolve/main/t5xxl_fp16.safetensors
FLUX VAE (335 MB) : https://huggingface.co/OwlMaster/realgg/resolve/main/ae.safetensors
Work ln progress.
Description
FAQ
Comments (4)
I have the right llama model (chatting with it in another tab), but the node can't see the images...
"I'm happy to help!
However, I don't see any image attached. Could you please share the image with me?"
https://imgbox.com/6vcpGms5
Do you have Ollama installed?https://ollama.com/download/windows
And FYI I couldnt see your pics didnt load?
@admajic I have that; it starts with Windows. And I also started a chat with it, just to be sure the model is loaded.
https://images2.imgbox.com/52/b8/6vcpGms5_o.png
https://images2.imgbox.com/2a/06/SJBCpRc7_o.png
And it must work, because the llama text generator does what it's supposed to do. Only the describe function is not working.
thank you, it's very cozy here
good job