github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #12004] Chat completion API throws errors for reasoning_effort set to false (gpt-oss) and 'minimal' level is not supported #7970

New Issue
Closed
opened 2026-04-12 20:09:19 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-12 20:09:19 -05:00
Owner
Copy Link

Originally created by @Cdany2001 on GitHub (Aug 21, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12004

What is the issue?

I'm running gpt-oss:20b and I encountered the following problems:

  • the OpenAI endpoint supports values such as minimal/low/medium/high but the Ollama version of the endpoint seems to not be fully compatible and it allows only the last 3 levels + the possibility to set it to true or false (Ollama response: "message": "invalid think value: \"minimal\" (must be \"high\", \"medium\", \"low\", true, or false)" )
  • when attempting to set it to false via the API request i get this error:
    { "error": { "message": "json: cannot unmarshal bool into Go struct field ChatCompletionRequest.reasoning_effort of type string", "type": "invalid_request_error", "param": null, "code": null } }
    To disable the thinking, are there other fields from the request that I should populate instead?

Relevant log output

OS

Linux, Docker

GPU

Nvidia

CPU

Intel

Ollama version

0.11.6

Originally created by @Cdany2001 on GitHub (Aug 21, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12004 ### What is the issue? I'm running gpt-oss:20b and I encountered the following problems: - the OpenAI endpoint supports values such as [minimal/low/medium/high](https://platform.openai.com/docs/api-reference/chat/create#chat_create-reasoning_effort) but the Ollama version of the endpoint seems to not be fully compatible and it allows only the last 3 levels + the possibility to set it to true or false `(Ollama response: "message": "invalid think value: \"minimal\" (must be \"high\", \"medium\", \"low\", true, or false)" )` - when attempting to set it to false via the API request i get this error: `{ "error": { "message": "json: cannot unmarshal bool into Go struct field ChatCompletionRequest.reasoning_effort of type string", "type": "invalid_request_error", "param": null, "code": null } }` To disable the thinking, are there other fields from the request that I should populate instead? ### Relevant log output ```shell ``` ### OS Linux, Docker ### GPU Nvidia ### CPU Intel ### Ollama version 0.11.6
GiteaMirror added the bug label 2026-04-12 20:09:19 -05:00
GiteaMirror commented 2026-04-12 20:09:20 -05:00
Author
Owner
Copy Link

@Cdany2001 commented on GitHub (Aug 28, 2025):

I'll close this since:

  • yes the OpenAI endpoints allow "minimal" as a rasoning_effort value, but it is not supported by Gpt-Oss
  • from the code, I see that the message that states that the think value can be a bool is coming from a class that funnels in different attributes from different models/parameters, while the error that I received when i attempted to set it to false was coming from a class that was higher in the hierarchy and described how the chat completion request should look like (and so allowing only strings) => tbh, an improvement of the error describing the allowed values would be welcomed
<!-- gh-comment-id:3233660393 --> @Cdany2001 commented on GitHub (Aug 28, 2025): I'll close this since: - yes the OpenAI endpoints allow "minimal" as a rasoning_effort value, but it is not supported by Gpt-Oss - from the code, I see that the message that states that the think value can be a bool is coming from a class that funnels in different attributes from different models/parameters, while the error that I received when i attempted to set it to false was coming from a class that was higher in the hierarchy and described how the chat completion request should look like (and so allowing only strings) => tbh, an improvement of the error describing the allowed values would be welcomed
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#7970
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?