Adapt Comfyui Python Script to use as an API
Budget: ₹1,500 – ₹12,500 INR
Hey there! I am looking for someone who can take the example of an API with ComfyUI and adapt it to my needs.
Here is the example - https://github.com/comfyanonymous/ComfyUI/blob/master/script_examples/websockets_api_example_ws_images.py.
What I need to happen is that the JSON which is the ComfyUI workflow is dynamic so I can provide that in an API file / raw JSON to trigger the start of a workflow.
I would also like the websocket to be hardcoded to a channel that I operate and that it is connected by default when the machine is on so that everytime that the script is run, it does not try to connect again.
The final part is packaging this up with instructions so I know how to turn it on and run it on my local machine and then how I would call the endpoint. I have familiarity with using Ngrok for Automatic1111 in AI image generation so I plan to use Ngrok again to turn this locally into something I can call from elsewhere.
An even better solution for me which I am open to would be me defining a webhook URL which is hit when an image is complete. If possible to do this and completely sidestep the websocket needs then that would be something that I am interested in.
Ideally for the image response, it would also send me a job ID or something so we can track the image response to a job id from the initial call to the output.
Looking for someone who has experience with AI image generation, python and making all of these systems work! Not looking for agencies.
Please do not response with a cookie cutter AI response, these will be automatically rejected.
The project will be classed as complete when I have it running on my local machine and have sent a successful image request and got a successful image response
Pls send link of sample work
Here is the example - https://github.com/comfyanonymous/ComfyUI/blob/master/script_examples/websockets_api_example_ws_images.py.
What I need to happen is that the JSON which is the ComfyUI workflow is dynamic so I can provide that in an API file / raw JSON to trigger the start of a workflow.
I would also like the websocket to be hardcoded to a channel that I operate and that it is connected by default when the machine is on so that everytime that the script is run, it does not try to connect again.
The final part is packaging this up with instructions so I know how to turn it on and run it on my local machine and then how I would call the endpoint. I have familiarity with using Ngrok for Automatic1111 in AI image generation so I plan to use Ngrok again to turn this locally into something I can call from elsewhere.
An even better solution for me which I am open to would be me defining a webhook URL which is hit when an image is complete. If possible to do this and completely sidestep the websocket needs then that would be something that I am interested in.
Ideally for the image response, it would also send me a job ID or something so we can track the image response to a job id from the initial call to the output.
Looking for someone who has experience with AI image generation, python and making all of these systems work! Not looking for agencies.
Please do not response with a cookie cutter AI response, these will be automatically rejected.
The project will be classed as complete when I have it running on my local machine and have sent a successful image request and got a successful image response
Pls send link of sample work