Help with Stable-diffusion-Automatic(1111) webui
I could not find IP-Adapter control type in controlnet on webui ,webui is not showing IP-Adapter option to select ,kindly help me in that
52 Replies
Which template?
runpod/stable-diffusion:web-ui-10.2.1
That template is very outdated and has a very old version of A1111 and ControlNet, I don't recommend using it if you want to use things like IP-Adapter. Use this community template instead, it has latest A1111 and latest ControlNet.
Does this have UI and provide a URL to use it in API?
Yes, it includes
--api command line argument to enable the API.
Simply add /docs to the end of the URL and you can access the API docs.I can't find the above mentioned template. Can you please provide the link?
Click "Expore" in the left hand nav, then type A1111 into the search box
I did same ,it shows No results found
just put A1111
Actually I think yours isn't loading properly because it seems to always show vllm and yours isn't
Try see if you can find a link to the template on the Github page
https://github.com/ashleykleynhans/a1111-docker?tab=readme-ov-file
GitHub
GitHub - ashleykleynhans/a1111-docker: Docker image for Stable Diff...
Docker image for Stable Diffusion WebUI with ControlNet, After Detailer, Deforum and ReActor extensions - ashleykleynhans/a1111-docker
I find the link but its not taking me to the template
You have to select secure or community cloud, then select a GPU type, then you will see the template before deploying the pod.
I don't know why RunPod changed it, it was more useful the way it was previously where you could see the template before even selecting which cloud type you want to use.
Marcus Please help I don't find this template in any of the cloud whether community or secure cloud
Looks like this is the link, from the Github, did you try it?
https://www.runpod.io/console/deploy?template=ts8ze6urzh
Yes, the github link is not working for me
Your link takes me to the same cloud page which github link does ,no difference
It doesn't show the template like it used to
First select community or secure cloud, then a GPU
Step 1
Step 2
Step 3, scroll down and you will see it here
I selected RTX A4000 and it gives me template I have deployed it ,its showing running but none of the port is working and it is stuck on 97%
Its not stuck, wait for it to complete.
Alright
where shall I change command line arguments in order to use api,in linux file or windows file
Definitely not Widows, RunPod is Linux only.
This template already has API enabled, you do not need to change anything.
yes thanks
hey marcus?
are you there actually I am getting unexpected error which I was not getting 2 days before
RuntimeError: mat1 and mat2 must have the same dtype, but got Float and Half
I am getting this error
Also, I already have --no-half in my command line arguments, as recommended by some people on stack overflow
Where did you put it? In COMMANDLINE_ARGS inside webui-user.sh?
yes
I was having the same setting on friday and its was working fine
Must be a problem with pod or model then
So what should I do now ?
if this issues lies with model then It shouln't be working on Friday too
kindly tell me what should I do now
Try a different pod
can you suggest the the alternate of A1111 Stable Diffusion 1.9.4
ComfyUI
I want to use Automatic1111
no other UI as I have developed its API and add many custom functions in it ,can not go back and shift whole UI
I think you got a bad pod if it worked and then didn't, the template hasn't changed over the weekend.
so shall I change pod and keep same template?is that you suggesting?
so this template A1111 Stable Diffusion 1.9.4 available on another pod?
@Marcus Waiting fro your reply
I think for some templates it's must be reinstalled
Except if you use network storage
so what pod shall I use now ?
Your old one or try a new one
i want to use same template but it has problem
This?
What model were you using maybe try re-downloading the model again
Its not a template problem, you said it worked previously, and the template did not change, so how is it a template problem?
Either a problem with the pod or the model like I said.
Yeah
I would try a different pod because you said the model is also the same, so then that eliminates the model and basically must be something with the pod.
I am using cyberrealistic inpainting model ,vaepruned and lora models all are in .safetensors and controlnet ip adapter
My old pod was same but template was different it did not have ipadapter so I switch my template but now pod has problems
That's what I am asking its is a problem of pod so what pod should I use now with same template
Yes should be
Try with the same template
Or different maybe is fine if it doesn't works
Now UI is working fine but api has problem Api IS GIVING RUN TIME ERROR
What runtime error
gateway time out
Never seen it before
Any text logs or screenshot of the error
You will get gateway timeout from cloudflare if the request takes more than 100s to complete
Rather change port to TCP if you want to have long running API requests
Ah cloudflare
Yeah there's a timeout if your request is too long