AI Lab
AI Lab
> i solved this problem, > would you remove that 'save image' node and redeploy? > update and not redeploy made an error but after redeploy the error doesn't showed...
same here, any solution?
Thank you for your detailed report and thorough diagnosis. We’ve reviewed your findings, and it’s likely that the initial delay is due to llama-cpp-python initializing or compiling the model during...
what is your tranformers version and have you install the requirments.txt?
The error means the model didn’t finish loading correctly. Please update both ComfyUI and ComfyUI-JoyCaption to the latest version, then restart ComfyUI. ``` pip install -r requirements.txt ``` If it...
Thanks for the detailed report! This seems related to Python 3.12 and the latest PyTorch nightly build. I haven’t tested JoyCaption with that combination yet. Could you please try with...
Testing occurs on CUDA 12.8 and PyTorch 2.8 python 3.12 execution is smooth.
The message output in your log is normal — it just means no custom models were found. If your concern is that JoyCaption sometimes produces self-questioning lines like “What is...
Have you attempted the llama_cpp_python installation guide located at: https://github.com/1038lab/ComfyUI-JoyCaption/blob/main/llama_cpp_install/llama_cpp_install.md Following that, execute the 'llama_cpp_install.py' script within your ComfyUI environment using the command: ``` .\comfyui\python_embeded\python.exe llama_cpp_install.py ``` This operation may...
Thanks for pointing out the bug. We'll try to fix it in an upcoming update.