How do you use the custom model from the network volume?

[kilimchoi]

I was wondering how you make an API request to use the custom model from the network volume. It seems that if we use

{"input": ..., "4":{
  "inputs":{
     "ckpt_name":"realistic_vision_v5.1.safetensors"},
      "class_type":"CheckpointLoaderSimple"
   }
}

it just uses whatever model was included in the image. I added the network volume in the endpoint configuration on runpod btw.

[TimPietrusky]

@kilimchoi so this should work like this:

• The network volume is mounted as /runpod-volume into the serverless ComfyUI instance
• Add your model to the network volume, for example into a folder called checkpoints with the model realistic_vision_v5.1.safetensors
• Then it should look like this in your workflow:

{
  "input": {
     "4": {
       "inputs": {
         "ckpt_name":"/runpod-volume/checkpoints/realistic_vision_v5.1.safetensors"
       },
      "class_type":"CheckpointLoaderSimple"
     }
  }
}

[TimPietrusky]

@kilimchoi I'm reopening the issue, as I want to document this also in the README.

[dannykok]

It's kind of strange that I can load the custom model by using only the file name (without path). When I tried to use the full path of model file:

        "4": {
          "inputs": {
            "ckpt_name": "/runpod-volume/models/checkpoints/personaStyle_lite.safetensors"
          },
          "class_type": "CheckpointLoaderSimple"
        },

I got 400 bad request error. The logs show this:

Value not in list: ckpt_name: '/runpod-volume/models/checkpoints/personaStyle_lite.safetensors' not in ['personaStyle_lite.safetensors', 'sd_xl_base_1.0.safetensors']

I am using the timpietruskyblibla/runpod-worker-comfy:3.0.0-sdxl docker.

[TimPietrusky]

@dannykok @kilimchoi what I wrote is wrong, I totally forgot about the mapping that we are doing in extra_model_paths.yaml. So you can just specify the name of the model, you don't need to specify the path.

I will make sure to get the README updated to make this clear.

[albertogb9]

I tried with and without the path but the network volume isn't being recognized by the worker

[dannykok]

@albertogb9 you may want to check if the path you used exists in extra_model_paths.yaml.

[albertogb9]

@albertogb9 you may want to check if the path you used exists in extra_model_paths.yaml.

Already figured it out. I had installed comfyui in the network volume, so the path was "ComfyUI/models/..." whereas it should be just "models/...". So I just created another network volume with a pytorch template and I created the models folders from scratch and loaded the checkpoints. Now it works fine, thanks.

[IMParesqui]

Im getting the same error as above.
I tried creating a custom path for the model: 'checkpoints: models/checkpoint2/'
I created a netwok volume with the sdxl_lightning_4step_unet.safetensors inside checkpoint2.
I added the "inputs": {
"ckpt_name": "/runpod-volume/checkpoint/sdxl_lightning_4step_unet.safetensors"

Value not in list: ckpt_name: '/runpod-volume/checkpoint2/sdxl_lightning_4step_unet.safetensors' not in ['sdxl_lightning_4step_unet.safetensors']

Was the readme updated? How do I modify the 'extra_model_paths' in order to fix this?

Also, is this needed?
echo "Symlinking files from Network Volume"
ln -s /runpod-volume /workspace

[sirkian]

Im getting the same error as above. I tried creating a custom path for the model: 'checkpoints: models/checkpoint2/' I created a netwok volume with the sdxl_lightning_4step_unet.safetensors inside checkpoint2. I added the "inputs": { "ckpt_name": "/runpod-volume/checkpoint/sdxl_lightning_4step_unet.safetensors"

Value not in list: ckpt_name: '/runpod-volume/checkpoint2/sdxl_lightning_4step_unet.safetensors' not in ['sdxl_lightning_4step_unet.safetensors']

Was the readme updated? How do I modify the 'extra_model_paths' in order to fix this?

Also, is this needed? echo "Symlinking files from Network Volume" ln -s /runpod-volume /workspace

Have you tried removing "/runpod-volume/checkpoint/" from the ckpt_name path? That might solve your issue, because by this: ['sdxl_lightning_4step_unet.safetensors'], it looks like the model gets recognized.

Below got solved by using base image as serverless template (#60 )
I'm also having similar issues: - Value not in list: ckpt_name: 'Juggernaut_X_RunDiffusion.safetensors' not in ['sd_xl_base_1.0.safetensors'].

Custom nodes work fine from runpod-volume/custom_nodes/, but runpod-volume/models/checkpoints/ doesn't seem to work.

[IMParesqui]

I did, I tried with:
/runpod-volume/checkpoint2/sdxl_lightning_4step_unet.safetensors"
/checkpoint2/sdxl_lightning_4step_unet.safetensors"
/sdxl_lightning_4step_unet.safetensors"
sdxl_lightning_4step_unet.safetensors"

the first three returned the same error, and with the last one I got:

!!! Exception during processing !!! ERROR: Could not detect model type of: /runpod-volume/checkpoint2/sdxl_lightning_4step_unet.safetensors

By the way, I had some troubles acessing the folder "models/checkpoints" in the network volume, that's why I changed it's location. Maybe thats the reason it works for custom_nodes but not for the checkpoints.

[sirkian]

By the way, I had some troubles acessing the folder "models/checkpoints" in the network volume, that's why I changed it's location. Maybe thats the reason it works for custom_nodes but not for the checkpoints.

I also had trouble accessing models/checkpoints via jupyter folders after downloading first model, but I was able to % ls models/checkpoints to confirm that it is in there.

No idea what could cause your issue though.