Thank you for your work, i got it running yesterday. Runpod charges 0.02$ per hour for storage even when i'm not running the runpod. Is this normal or is there a way to avoid that?
Yes this is normal as you are using storage. You can delete the pod and start a fresh one each time if you don't want to pay for storage. Sometimes even when I want to pay for storage I can't start with the same GPU so the pod is not very useful. I always delete and start fresh one each time.
@@weirdwonderfulaiartyou can also save a snapshot of your work setup. Then just store that. It’ll greatly reduce size to save money. Alternatively you can save that snapshot to a free cloud storage if you won’t be returning for a while. I think Daniel has the right idea though.
I must be blind but i cannot see any notebook in the workspace directory, only in the comfy directory and the notebook does not have the workflow in the video nor a flux model download. has this been removed? or can you please direct me to it?
you can view any log using Tail command eg. tail -f /workspace/logs/comfyui.log -f allows you to follow a file and updates as it changes you can see GPU utilisation using nvidia-smi in terminal and you can use it with watch command to monitor continously eg. watch -d -n 1 nvidia-smi -n, --interval Allows you to specify the interval between output updates. -d, --differences Highlights the differences between output updates.
can you make a video on how i can run omnigen on this? runpod io .. thanks
I really want you to continue on this is awesome
Will produce more soon. Thanks for the encouragement
@@weirdwonderfulaiart I’m here for ad
Thank you for your work, i got it running yesterday. Runpod charges 0.02$ per hour for storage even when i'm not running the runpod. Is this normal or is there a way to avoid that?
Yes this is normal as you are using storage. You can delete the pod and start a fresh one each time if you don't want to pay for storage.
Sometimes even when I want to pay for storage I can't start with the same GPU so the pod is not very useful.
I always delete and start fresh one each time.
@@weirdwonderfulaiart Okay, thank you. Makes sense
@@weirdwonderfulaiartyou can also save a snapshot of your work setup. Then just store that. It’ll greatly reduce size to save money. Alternatively you can save that snapshot to a free cloud storage if you won’t be returning for a while. I think Daniel has the right idea though.
Also how do I add custom checkpoints/tensors?
Download them into the models folder
Nice job explaining this. Would like a deep dive when you have the time. Thank you so much for sharing! Subbed and liked
Awesome, thank you!
Thank you SO much for this!!
You're so welcome!
Can you please create a tutorial that can do batch of prompts? I mean multiple prompts queued one by one in single click?
You can easily do this by adding the DynamicPrompts Custom Nodes to generate multiple prompts and batch them up in comfyui
thanks for the video, how do i install civit ai models?
I have a Jupyter notebook in the resources link which will download the models
I must be blind but i cannot see any notebook in the workspace directory, only in the comfy directory and the notebook does not have the workflow in the video nor a flux model download. has this been removed? or can you please direct me to it?
Link in the description of the video to download the zip which contains the notebook.
When I create the huggingface token, it says 403 forbidden :(
I found the answer, I needed write access for my token. Thank you for providing the files!
You only need read Access as you are pulling the model down and not writing anything
@@DjSligs Did you put "< >" around the token? I have the same issue as you did but it's not solved by just inputing a write token. Any help?
How did you load the log viewer in the terminal and how can I view the resource utilization in the terminal?
you can view any log using Tail command eg. tail -f /workspace/logs/comfyui.log
-f allows you to follow a file and updates as it changes
you can see GPU utilisation using nvidia-smi in terminal and you can use it with watch command to monitor continously eg. watch -d -n 1 nvidia-smi
-n, --interval Allows you to specify the interval between output updates.
-d, --differences Highlights the differences between output updates.
can you teach how to install forge on runpod to use flux dev
What if forge? I've only ever used Automatic1111 and ComfyUI
How to add more lora?
Duplicate the load Lora node save then link it into the chain
@ sorry but i dont get it how to connect the node. Can you give me some examples?
In your notebook, where do I put my hugging face token?
Refer to the screenshot on the blog post HF Token
weirdwonderfulai.art/resources/cheap-solution-for-running-flux-dev-using-runpod/
How do I get your work flow?
Description has the link to workflow
Hello do you have any tutorials on using the Flux Realism but with your own face?
not yet, wait for ControlNets and IPAdapters to be available for Flux then it would be possible
@@weirdwonderfulaiartwhy can't we just use a LoRA node? I use a character LoRA fine in Comfy local, but can't get it installed in Runpod.
@@robdhayes is it possible to load checkoints and lora? please show this.
I don't see your notebook?
All resources are available on the blog which is linked in the description
@@weirdwonderfulaiart I checked on the blog post and I still can't find the url for the notebook. Can you point me to the right direction?
nvm, found it. didn't realize it was on the package url in the end. Thanks a lot WW, this was ultra helpful.