mirror of
https://github.com/open-webui/open-webui.git
synced 2026-07-10 23:55:15 -05:00
[GH-ISSUE #19908] feat: Supports OpenAI-Response format API and related parameters #19036
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/open-webui#19036
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @daxiass on GitHub (Dec 12, 2025).
Original GitHub issue: https://github.com/open-webui/open-webui/issues/19908
Check Existing Issues
Verify Feature Scope
Problem Description
Dear Author,
We have now entered the era of reasoning-model AI. However, our OpenWebUI still does not provide proper basic support for OpenAI’s reasoning models: we can’t view the reasoning process or the thinking time, and we also can’t configure verbosity.
At the moment, I’m only able to achieve this by hooking in an external function, which is very inconvenient. Therefore, I kindly request that you add support for OpenAI’s official Responses API.
Desired Solution you'd like
support for OpenAI’s official Responses API.
Alternatives Considered
No response
Additional Context
No response
@owui-terminator[bot] commented on GitHub (Dec 12, 2025):
🔍 Similar Issues Found
I found some existing issues that might be related to this one. Please check if any of these are duplicates or contain helpful solutions:
#15356 feat: Support OpenAI responses API to enable new model compatibility
by PeterDaveHello • Jun 27, 2025
#11758 feat: Support for OpenAI's new Responses API
by PierreMesure • Mar 17, 2025
#11893 feat: support openai responses api
by goguryeo1 • Mar 20, 2025
#12583 feat: Model Content Protocol
by fishcharlie • Apr 07, 2025
#10312 ApiEndpoint support open ai functions call
by mingyangzhu • Feb 19, 2025
Show 5 more related issues
#18575 feat: Support for OpenAI gpt-5-pro
by i-rme • Oct 24, 2025
#13101 feat: Add Support for Adjusting Reasoning Intensity in OpenAI-Compatible Mode
by L0stInFades • Apr 21, 2025
#9093 feat: support
include_reasoningparameterby coxde • Jan 29, 2025
#12682 feat: Support ReAct
by xlxxcc • Apr 10, 2025
#9300 Could OpenAI API Connections has thought process? the same as deepseek R1
by sixyang • Feb 03, 2025
💡 Tips:
This comment was generated automatically by a bot. Please react with a 👍 if this comment was helpful, or a 👎 if it was not.
@daxiass commented on GitHub (Dec 12, 2025):
I think the sheer number of voices calling for this further confirms the urgency of this need—and I'm not the only one who feels this way.
@Mcayear commented on GitHub (Dec 12, 2025):
OpenWebUI has deviated from its original commitment to "comply with the OpenAI API."
@Classic298 commented on GitHub (Dec 12, 2025):
@Mcayear @daxiass
Incorrect. Tim has said that Open WebUI will only support OpenAI Chat Completions API - aka the original OpenAI API that is also widely adopted where multiple providers (such as Google, Openrouter and even Anthropic) offer official production or at least Beta endpoints that are OpenAI completions API compatible.
The Responses API is a new API, that basically no provider supports except OpenAI themselves and much of the functionality is not even replicable the same way OpenAI provides it - i.e. server sided chat management, web search, the built-in function calling with server sided functions and more - all the extra quirks that the Responses API support are, in the way they are supported, exclusive to OpenAI's offering - thus making the Responses API something not every provider can just support as a universal standard like the chat completions API.
This was addressed MULTIPLE times in DOZENS of issues and discussions.
I urge you both to search for those before opening yet another issue.
As you can see there are many existing issues, the bot found them, and Tim clarified multiple times what I am just repeating here.
Asking for Responses API support is like asking for Anthropic API or Gemini API support. These are (semi) proprietary APIs that are exclusive to that provider, wheras the OpenAI completions API is the de-facto universal standard.