RunPod

R

RunPod

We're a community of enthusiasts, engineers, and enterprises, all sharing insights on AI, Machine Learning and GPUs!

Join

⚡|serverless

⛅|pods

Getting Available Values for Stable Diffusion API Parameters.

I am trying to figure out where to get the official Keys:Values for the API Parameters for the FASTAPI Stable Diffusion setup. For instance, For the Sampling Method, there is a drop down with values that you can clearly see and use in the Automatic 1111 WebUI, but how do I know what the official parameter is and the exact values available so that when I send a request via the API it sends the right value? In the swagger example it provides some of this, but clearly not all. And certainly does not provide the entire available values. Specifically, I am looking for the official parameter for selecting the Stable Diffusion Checkpoint, How to get the specific values available for the Sampling Method and Schedule type. In the Automatic 1111 webui screenshot - Are these the official Checkpoint model names to use? Including the string in the brackets after the name? By selecting the "/openapi.json" (screenshot) (small link at the very top of the API documentation page) It gives a HUGE JSON file which you can search through and get clues to what the parameter values are. Is this a good practice to find this information this way?...
No description

Which template to use?

Is there any advantage to using ashleykza's A1111 Stable Diffusion 1.9.4 over the official runpod template (unpod/stable-diffusion:web-ui-10.2.1)?

A100 OneTrainer stuck in downloading loop

Hi there, I am trying out the OneTrainer template and it seems to be broken for the A100 GPU's. After everything has been downloaded, it seems to go back to the "Extracting" part again and starts downloading again, like an infinite loop. This does not happen for a RTX4090 for example...
No description

OneTrainer

Hi there, I am trying out the OneTrainer template but how do I see the actual GUI? Since there is no web app. Is there some VNC connection?

Locking down web accessible items

So I have a stupid question, still trying to understand how all of this works. whenever I deploy a pod, for Text Generation Web UI and APIs I get a list of services that are available via http from the get go. I already added a key so the web interface is locked but I still need to lock down the file upload, visual studio code, and juniper notebook. Does anyone have a where I can read up on how to do this?...
No description

How to fetch more than 8 gpus on RunPod (2 nodes)

Hi, Usually RunPod provides max 8 gpus, how i can fetch more than 8 gpus?

Environment Variables are not set in "SSH over exposed TCP" for the root user

The template env variables are not available in the "SSH over exposed TCP" connection which is the root user. I am wondering if the reason for this is because they are only set for the "Basic SSH" user?

Environment variables missing

Hello, I am creating a pod with environment variables, but it doesn't seem to work. When I connect via SSH, and echo $ENV_VAR_NAME it prints nothing. Am I missing something? Also, printenv doesn't show the default environment variables from runpod, nor my added environment vars. I am using my own template, but the Docker image is build on top of an official runpod image....
Solution:
Its docker yeah, environment in linux is per user i guess, so when you login using ssh, your env's wont be there because the docker container starts your application as a different user

GPU Memory Used Issue

Can anyone please explain to me why I am using 93% of my memory while running nothing? I imported my comfyUI workflow but not actively running the flow Are there any way I can allocate it to reduce it?...
No description

Avast Antivirus detects Runpod as a Trojan Virus

Hello For some reason my Anti-Virus detects api.runpod.io as a Trojan Horse virus, which renders the service unusable. The Alert pops up after selecting a GPU to use, then clicking on the "Change Template" button to select a Pod Template....
No description

vulkan

the GPU-accelerated remote desktop, where can i find details on the GPU? i see openGL is installed, can i install vulkan as well?

how do I add storage to a existing running pod?

how do I add storage to a existing running pod?

Quick Question for SSH connection

Quick question for you guys. I've executed train.py from my remote container via VSCode ssh connection. I don't know why that execution killed or stopped? right after wifi disconnection. Is there anyone help me out? I didn't stop my pod.
Solution:
i prefer to use tmux

Failed to initialize NVML: Unknown Error

(compress) root@1908bfec7b85:/workspace# nvidia-smi
Failed to initialize NVML: Unknown Error
Failed to initialize NVML: Unknown Error
...

Runpod fast Stable Diffusion does not working

So I started pod today, but it failed. I think it is related to the Last Ben's repo. Here are part of the loggs: 2024-07-19T10:54:54.462970632+02:00 Starting Nginx service... 2024-07-19T10:54:54.474794604+02:00 * Starting nginx nginx 2024-07-19T10:54:54.482832995+02:00 ...done. 2024-07-19T10:54:54.483416353+02:00 Running pre-start script......

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

Help with Connections from my website

I want to create a web application using a pod but I don't know how I can make connections from my website if every time I restart the pod the IP and TCP port mapping change. Any experience with this? Any tutorial? Thank you so much!
Solution:
IP and TCP port can be accessed from env variable so you can figure something out to pass them into your other code this...

Cannot start fast stable diffusion notebook

It was fine this morning but now im getting error messages when starting the pod such as invalid username and password. When following the link to the LastBen repo on huggingface it seems to no longer exist.
Solution:
Template has been discontinued/abandoned. RunPod should delist it.

how to stop scripts from replacing my configuration

I have a version of ComfyUI together with a version of python and it's dependendencies in a direcotry named /workspace on a storage volume. Everytime this pre_start script runs it wipes out my lib and starts over. I tried modifying the script but it gets regenerated every time I restart the pod. How to stop this?
No description