What happened?
After launching ai-studio, I switched to the Hi-Dream lane and tried to generate an image by typing a simple prompt but I was met with a "Image generation failed: ComfyUI generation error" message.
Note: I was unable to complete step 2 of the onboarding steps after running setup-ai-studio.sh.
2. On the 'Studio' function (gear icon), set the 'browser_base' valve to" echo " http://$LANIP:8189 so returned media links open from your browser."
There was no "Studio" function available in the OpenWebUI interface and there were no valves listed under the Function/Studio section of the Controls panel.
Steps to reproduce
* bash scripts/gpu-mode.sh ai-studio
* Open OpenWebUI interface and select Hi-Dream lane
* Type in arbitrary prompt and send
* Notice error
Rig report — paste contents of bash scripts/report.sh [--verify | --soak | --bench | --full]
═══ Switching to AI-STUDIO mode (image · video · audio · voice) ═══
Starting: ComfyUI :8188 (both GPUs) + director :8090 + gallery/orchestrator/image-shim/tts + Open WebUI
Stopping: all GPU-bound LLM serving (Qwen + Gemma + DiffusionGemma)
▼ Stopping 27b-dual-mtp...[sudo: authenticate] Password:
done
▼ Stopping 27b-dual-dflash...done
▼ Stopping 27b-dflash-noviz...done
▼ Stopping 27b-dual-turbo...done
▼ Stopping deckard-40b...done
▼ Stopping 35b-a3b-dual...done
▼ Stopping gemma-mtp...done
▼ Stopping gemma-int8...done
▼ Stopping gemma-12b...done
▼ Stopping diffusiongemma-26b-a4b...done
[+] up 1/1ng comfyui...
✔ Container comfyui Started 0.5s
done
[+] up 1/1ng studio-director (:8090, GPU0)...
✔ Container studio-director Running 0.0s
done
[+] up 1/1ng studio-gallery (:8189)...
✔ Container studio-gallery Running 0.0s
done
[+] Building 1.7s (11/11) FINISHED
=> [internal] load local bake definitions 0.0s
=> => reading from stdin 606B 0.0s
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 335B 0.0s
=> [internal] load metadata for docker.io/library/python:3.12-slim 1.2s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 2B 0.0s
=> [1/4] FROM docker.io/library/python:3.12-slim@sha256:6c4dd321d176d61ea848dc8c73a4f7dbae8f70e0ee48bb411ea2f045b599fa8e 0.0s
=> => resolve docker.io/library/python:3.12-slim@sha256:6c4dd321d176d61ea848dc8c73a4f7dbae8f70e0ee48bb411ea2f045b599fa8e 0.0s
=> [internal] load build context 0.0s
=> => transferring context: 37B 0.0s
=> CACHED [2/4] RUN apt-get update && apt-get install -y --no-install-recommends ffmpeg && rm -rf /var/lib/apt/lists/* 0.0s
=> CACHED [3/4] COPY orchestrator.py /app/orchestrator.py 0.0s
=> CACHED [4/4] WORKDIR /app 0.0s
=> exporting to image 0.1s
=> => exporting layers 0.0s
=> => exporting manifest sha256:b757d55365635473a12766fe713938aee485cd6d00dbcd4e4e5cd08029ce95b3 0.0s
=> => exporting config sha256:596c1fc0c5d9ad6438675c585251d4fac7a66e16aeddb3d0e5fc33c6e4f0e991 0.0s
=> => exporting attestation manifest sha256:0140116ed069cf527d1b5dfcfead10db231732d79f802a6c07308d988a1e6912 0.0s
=> => exporting manifest list sha256:51eedf248ff76e9be8ee3296ef4d08d960dc808469425cc6608d57820734f631 0.0s
=> => naming to docker.io/library/studio-orchestrator:latest 0.0s
=> => unpacking to docker.io/library/studio-orchestrator:latest 0.0s
=> resolving provenance for metadata file 0.0s
[+] up 2/2
✔ Image studio-orchestrator:latest Built 1.7s
✔ Container studio-orchestrator Started 10.3s
done
[+] Building 0.7s (11/11) FINISHED
=> [internal] load local bake definitions 0.0s
=> => reading from stdin 592B 0.0s
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 192B 0.0s
=> [internal] load metadata for docker.io/library/python:3.12-slim 0.3s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 2B 0.0s
=> [1/4] FROM docker.io/library/python:3.12-slim@sha256:6c4dd321d176d61ea848dc8c73a4f7dbae8f70e0ee48bb411ea2f045b599fa8e 0.1s
=> => resolve docker.io/library/python:3.12-slim@sha256:6c4dd321d176d61ea848dc8c73a4f7dbae8f70e0ee48bb411ea2f045b599fa8e 0.0s
=> [internal] load build context 0.0s
=> => transferring context: 29B 0.0s
=> CACHED [2/4] RUN pip install --no-cache-dir aiohttp==3.10.11 0.0s
=> CACHED [3/4] WORKDIR /app 0.0s
=> CACHED [4/4] COPY shim.py /app/shim.py 0.0s
=> exporting to image 0.1s
=> => exporting layers 0.0s
=> => exporting manifest sha256:1ee60122df59c6c4d008b1984b0bf8b352c15a8439431336a979204b3347653f 0.0s
=> => exporting config sha256:4c76470dc1f87cf962b59518788a89d5f55126cc551b2c1adf5973eb0d4cb652 0.0s
=> => exporting attestation manifest sha256:1710390468de23c5f89dd98fefb2256c82db5fa86c9efd910a02db43248a6618 0.0s
=> => exporting manifest list sha256:aab5fb8e972370ee60c4db0a22079f01aafec51a7ce736d8a0189c64d9fcf5d6 0.0s
=> => naming to docker.io/library/studio-image-shim:latest 0.0s
=> => unpacking to docker.io/library/studio-image-shim:latest 0.0s
=> resolving provenance for metadata file 0.0s
[+] up 2/2
✔ Image studio-image-shim:latest Built 0.7s
✔ Container studio-image-shim Started 0.3s
done
[+] Building 0.7s (12/12) FINISHED
=> [internal] load local bake definitions 0.0s
=> => reading from stdin 543B 0.0s
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 517B 0.0s
=> [internal] load metadata for docker.io/library/python:3.12-slim 0.3s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 2B 0.0s
=> [1/5] FROM docker.io/library/python:3.12-slim@sha256:6c4dd321d176d61ea848dc8c73a4f7dbae8f70e0ee48bb411ea2f045b599fa8e 0.1s
=> => resolve docker.io/library/python:3.12-slim@sha256:6c4dd321d176d61ea848dc8c73a4f7dbae8f70e0ee48bb411ea2f045b599fa8e 0.0s
=> [internal] load build context 0.0s
=> => transferring context: 28B 0.0s
=> CACHED [2/5] RUN apt-get update && apt-get install -y --no-install-recommends espeak-ng ffmpeg libsndfile1 && rm -rf /var/lib/apt/lists/* 0.0s
=> CACHED [3/5] RUN pip install --no-cache-dir kokoro-onnx soundfile aiohttp 0.0s
=> CACHED [4/5] WORKDIR /app 0.0s
=> CACHED [5/5] COPY tts.py /app/tts.py 0.0s
=> exporting to image 0.1s
=> => exporting layers 0.0s
=> => exporting manifest sha256:0ca8e525256febf5ea214c5c5ae49eb76c954f456d2fbaed06ab8cbc211f1e5c 0.0s
=> => exporting config sha256:544dcd89338a3b82113d10a357423b952a869974e6a6b91557d3cb35e56a7f32 0.0s
=> => exporting attestation manifest sha256:457bcb392a8afc51f741bbe09439f8a1c22289cd86822810a8865926f4587588 0.0s
=> => exporting manifest list sha256:8d5d40923ba99f9035ba2fa1cc7926fae1d659143cf157558ba7aeca0b5be94e 0.0s
=> => naming to docker.io/library/studio-tts:latest 0.0s
=> => unpacking to docker.io/library/studio-tts:latest 0.0s
=> resolving provenance for metadata file 0.0s
[+] up 2/2
✔ Image studio-tts:latest Built 0.7s
✔ Container studio-tts Started 0.4s
done
[+] Building 1.5s (14/14) FINISHED
=> [internal] load local bake definitions 0.0s
=> => reading from stdin 592B 0.0s
=> [internal] load build definition from Dockerfile 0.0s
=> => transferring dockerfile: 1.99kB 0.0s
=> [internal] load metadata for docker.io/pytorch/pytorch:2.7.0-cuda12.8-cudnn9-runtime 1.0s
=> [internal] load .dockerignore 0.0s
=> => transferring context: 2B 0.0s
=> [1/7] FROM docker.io/pytorch/pytorch:2.7.0-cuda12.8-cudnn9-runtime@sha256:7db0e1bf4b1ac274ea09cf6358ab516f8a5c7d3d0e02311bed445f7e236a5d80 0.0s
=> => resolve docker.io/pytorch/pytorch:2.7.0-cuda12.8-cudnn9-runtime@sha256:7db0e1bf4b1ac274ea09cf6358ab516f8a5c7d3d0e02311bed445f7e236a5d80 0.0s
=> [internal] load build context 0.0s
=> => transferring context: 31B 0.0s
=> CACHED [2/7] RUN apt-get update && apt-get install -y --no-install-recommends git ffmpeg libsndfile1 && rm -rf /var/lib/apt/lists/* 0.0s
=> CACHED [3/7] RUN git clone https://github.com/Saganaki22/ComfyUI-Step_Audio_EditX_TTS.git /opt/step && git -C /opt/step checkout d8eddc6618fdbe14a10cc99d13360 0.0s
=> CACHED [4/7] RUN pip install --no-cache-dir -r /opt/step/requirements.txt 0.0s
=> CACHED [5/7] RUN pip install --no-cache-dir aiohttp einops ffmpeg-python rotary_embedding_torch 0.0s
=> CACHED [6/7] COPY server.py /opt/server.py 0.0s
=> CACHED [7/7] WORKDIR /opt 0.0s
=> exporting to image 0.1s
=> => exporting layers 0.0s
=> => exporting manifest sha256:9c555fabeac51aeac8f757f0c2aa34a581902463463b0b2696f4b49b0a544b98 0.0s
=> => exporting config sha256:1ef190b2f8a5904bb798a14c6cc77191de69a7bfea4a1e69ac7254a8a7ad8f5a 0.0s
=> => exporting attestation manifest sha256:38a739e057a6e1f6e5044a179067259658e32f81c097612cc2e7366a999ed6b8 0.0s
=> => exporting manifest list sha256:8b6ff055b050462ad3144fc88c6e9fd5f7439b9ebe451dd0459bf3ae5c08f2ae 0.0s
=> => naming to docker.io/library/studio-step-voice:latest 0.0s
=> => unpacking to docker.io/library/studio-step-voice:latest 0.0s
=> resolving provenance for metadata file 0.0s
[+] up 2/2
✔ Image studio-step-voice:latest Built 1.6s
✔ Container studio-step-voice Started 0.5s
done
[+] up 1/1ng openwebui...
✔ Container open-webui Running 0.0s
done
[+] up 1/1ng litellm...
✔ Container litellm Running 0.0s
done
[+] up 1/1ng qdrant...
✔ Container qdrant Running 0.0s
done
[+] up 1/1ng searxng...
✔ Container searxng Running 0.0s
done
AI-studio mode active. — one scene; pick the lane in Open WebUI.
Open WebUI: http://192.168.86.33:8080 (image · video · music · SFX · voice lanes)
Gallery: http://192.168.86.33:8189 (all generated media; survives ComfyUI down)
ComfyUI: http://192.168.86.33:8188 (full node graph / control)
First ComfyUI boot can take a few min (clones + node deps). Video DiT splits across both 3090s (DisTorch); image/audio lanes run on GPU0 beside the director.
GPU-mutex with the dual-card LLMs. Premium voice (step-audio-editx) is on-demand on GPU1 — mutually exclusive with an active video render.
Tail: sudo docker logs -f comfyui
Anything else?
#
What happened?
After launching ai-studio, I switched to the Hi-Dream lane and tried to generate an image by typing a simple prompt but I was met with a "Image generation failed: ComfyUI generation error" message.
Note: I was unable to complete step 2 of the onboarding steps after running setup-ai-studio.sh.
2. On the 'Studio' function (gear icon), set the 'browser_base' valve to" echo " http://$LANIP:8189 so returned media links open from your browser."There was no "Studio" function available in the OpenWebUI interface and there were no valves listed under the Function/Studio section of the Controls panel.
Steps to reproduce
Rig report — paste contents of
bash scripts/report.sh [--verify | --soak | --bench | --full]Anything else?
#