Currently, open-webui supports Ollama/OpenAI API Integration: Effortlessly integrate OpenAI-compatible APIs for versatile conversations alongside Ollama models. Customize the OpenAI API URL to link with LMStudio, GroqCloud, Mistral, OpenRouter, and more.
That's great. And I know other model providers also want to cooperate with open-webui to create easy, stable, and affordable services to the community. How can they add their api to the open-webui? Is there any instruction to do it? Thank you.
Originally created by @elricwan on GitHub (Mar 4, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/11162
# Feature Request
Currently, open-webui supports Ollama/OpenAI API Integration: Effortlessly integrate OpenAI-compatible APIs for versatile conversations alongside Ollama models. Customize the OpenAI API URL to link with LMStudio, GroqCloud, Mistral, OpenRouter, and more.
That's great. And I know other model providers also want to cooperate with open-webui to create easy, stable, and affordable services to the community. How can they add their api to the open-webui? Is there any instruction to do it? Thank you.
Use LiteLLM or use any of the many existing pipeline implementations to integrate other providers or build your own pipeline. Find pipelines on openwebui.com/filters and search for stuff like anthropic or google or whatever you want - or use LiteLLM
I think Tim has stated in the past, that he does not intend to support anything that doesn't confine with the OpenAI API standard or Ollama. So anything else you need to integrate yourself using a pipeline or things like LiteLLM or portkey.
Or for self-hosted stuff / personal stuff you can also check out openrouter (website) which also gives you access to nearly all models with a single OpenAI compatible API and key.
<!-- gh-comment-id:2698362648 -->
@Classic298 commented on GitHub (Mar 4, 2025):
Use LiteLLM or use any of the many existing pipeline implementations to integrate other providers or build your own pipeline. Find pipelines on openwebui.com/filters and search for stuff like anthropic or google or whatever you want - or use LiteLLM
I think Tim has stated in the past, that he does not intend to support anything that doesn't confine with the OpenAI API standard or Ollama. So anything else you need to integrate yourself using a pipeline or things like LiteLLM or portkey.
Or for self-hosted stuff / personal stuff you can also check out openrouter (website) which also gives you access to nearly all models with a single OpenAI compatible API and key.
Thank you for quick response! Currently the link cannot open, I will check later. Also, the model provider use the OpenAI API standard, just like together ai, and openrouter.
If I understand correctly, in order to integrate the model to the open-webui, the first step is to integrate the model api to the LiteLLM, then, the second step is to follow the link instruction, right?
<!-- gh-comment-id:2698421109 -->
@elricwan commented on GitHub (Mar 4, 2025):
Thank you for quick response! Currently the link cannot open, I will check later. Also, the model provider use the OpenAI API standard, just like together ai, and openrouter.
If I understand correctly, in order to integrate the model to the open-webui, the first step is to integrate the model api to the LiteLLM, then, the second step is to follow the link instruction, right?
Yeah if you wanna use LiteLLM, follow their docs for a quick setup (its pretty easy) and then just add LiteLLM API endpoint and api key to OpenWebUI and done
<!-- gh-comment-id:2698430273 -->
@Classic298 commented on GitHub (Mar 4, 2025):
Yeah if you wanna use LiteLLM, follow their docs for a quick setup (its pretty easy) and then just add LiteLLM API endpoint and api key to OpenWebUI and done
Just a follow up question, if we follow the steps and add LiteLLM API endpoint and api key to OpenWebUI, could the community of open-webui use it too? Or it is just for us.
<!-- gh-comment-id:2698434474 -->
@elricwan commented on GitHub (Mar 4, 2025):
Just a follow up question, if we follow the steps and add LiteLLM API endpoint and api key to OpenWebUI, could the community of open-webui use it too? Or it is just for us.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Originally created by @elricwan on GitHub (Mar 4, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/11162
Feature Request
Currently, open-webui supports Ollama/OpenAI API Integration: Effortlessly integrate OpenAI-compatible APIs for versatile conversations alongside Ollama models. Customize the OpenAI API URL to link with LMStudio, GroqCloud, Mistral, OpenRouter, and more.
That's great. And I know other model providers also want to cooperate with open-webui to create easy, stable, and affordable services to the community. How can they add their api to the open-webui? Is there any instruction to do it? Thank you.
@Classic298 commented on GitHub (Mar 4, 2025):
Use LiteLLM or use any of the many existing pipeline implementations to integrate other providers or build your own pipeline. Find pipelines on openwebui.com/filters and search for stuff like anthropic or google or whatever you want - or use LiteLLM
I think Tim has stated in the past, that he does not intend to support anything that doesn't confine with the OpenAI API standard or Ollama. So anything else you need to integrate yourself using a pipeline or things like LiteLLM or portkey.
Or for self-hosted stuff / personal stuff you can also check out openrouter (website) which also gives you access to nearly all models with a single OpenAI compatible API and key.
@elricwan commented on GitHub (Mar 4, 2025):
Thank you for quick response! Currently the link cannot open, I will check later. Also, the model provider use the OpenAI API standard, just like together ai, and openrouter.
If I understand correctly, in order to integrate the model to the open-webui, the first step is to integrate the model api to the LiteLLM, then, the second step is to follow the link instruction, right?
@Classic298 commented on GitHub (Mar 4, 2025):
Yeah if you wanna use LiteLLM, follow their docs for a quick setup (its pretty easy) and then just add LiteLLM API endpoint and api key to OpenWebUI and done
@elricwan commented on GitHub (Mar 4, 2025):
Just a follow up question, if we follow the steps and add LiteLLM API endpoint and api key to OpenWebUI, could the community of open-webui use it too? Or it is just for us.
@Classic298 commented on GitHub (Mar 4, 2025):
what? anyone can use LiteLLM in combination with openwebui. :)