Stable Diffusion RuntimeError Cuda Error

General
1

How do I solve this error when I try to do anything with Stable Diffusion? Doing this on a different Linux machine I would have tried adifferent PyTorch, but I’m still figuring Olares out.

RuntimeError: CUDA error: no kernel image is available for execution on the device CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1. Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

2

I’ve gotten things working by moving to cpu only with a more limited YAML setup. I’ll have to keep playing and see if I can get PyTorch working with RTX 5090

 containers:
    - name: sdwebuis
      image: docker.io/beclab/kldtks-sdwebui:v1.10.1-pii-t3
      command:
        - sh
        - '-c'
        - >-
          pip config set global.index-url
          https://pypi.joinolares.cn/root/olares3/+simple/; pip config set
          global.extra-index-url ""; chown -R 1000:1000
          /stable-diffusion-webui; python -u
          /stable-diffusion-webui/webui.py --listen --port 7860 
          --allow-code --enable-insecure-extension-access --api --no-hashing
          --gradio-queue --use-cpu all --no-half --precision full
      ports:
        - containerPort: 7860
          protocol: TCP
      env:
        - name: CLI_ARGS
          value: >-
            --allow-code --enable-insecure-extension-access --api
            --no-hashing --gradio-queue --use-cpu all --no-half --precision
            full
        - name: PGID
          value: '1000'
        - name: PUID
          value: '1000'
        - name: HOME
          value: /root
        - name: NVSHARE_DEBUG
          value: '1'
        - name: CUDA_LAUNCH_BLOCKING
          value: '1'