[GH-ISSUE #10954] When using the ollama tag in version 0.9.0, the API call for qwen3 is invalid with the parameter 'think' #32975

Closed
opened 2026-04-22 14:59:28 -05:00 by GiteaMirror · 3 comments
Owner

Originally created by @yoin528 on GitHub (Jun 3, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10954

What is the issue?

121/10000

实时翻译
I saw that the latest version 0.9.0 solved the thinking mode of deepseek and qwen3, and immediately updated ollama and conducted testing. However, during the testing process, I found that when using qwen3:4b to call API interfaces, adding the think parameter still enabled the thinking mode, and there was no label release indicating that thinking would be removed.

Relevant log output


OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @yoin528 on GitHub (Jun 3, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10954 ### What is the issue? 121/10000 实时翻译 I saw that the latest version 0.9.0 solved the thinking mode of deepseek and qwen3, and immediately updated ollama and conducted testing. However, during the testing process, I found that when using qwen3:4b to call API interfaces, adding the think parameter still enabled the thinking mode, and there was no label release indicating that thinking would be removed. ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-04-22 14:59:28 -05:00
Author
Owner

@yoin528 commented on GitHub (Jun 3, 2025):

Image

Image

The above is the detailed information I requested

<!-- gh-comment-id:2933214211 --> @yoin528 commented on GitHub (Jun 3, 2025): ![Image](https://github.com/user-attachments/assets/23b75548-0377-42df-89c9-ed84bd434781) ![Image](https://github.com/user-attachments/assets/848365c4-6020-4e47-905a-1d34b9a7d13b) The above is the detailed information I requested
Author
Owner

@jmorganca commented on GitHub (Jun 3, 2025):

Hi @yoin528 qwen3 was updated to include thinking – would it be possible to re-download with ollama pull qwen3:4b and ollama pull qwen3:8b (should be near-instant)

<!-- gh-comment-id:2933222795 --> @jmorganca commented on GitHub (Jun 3, 2025): Hi @yoin528 `qwen3` was updated to include thinking – would it be possible to re-download with `ollama pull qwen3:4b` and `ollama pull qwen3:8b` (should be near-instant)
Author
Owner

@yoin528 commented on GitHub (Jun 3, 2025):

Hi 已更新以包括思考 - 是否可以使用 and 重新下载(应该几乎是即时的)qwen3``ollama pull qwen3:4b``ollama pull qwen3:8b

Thank you, it will be fine after updating the model qwen3:4b and qwen3:8b

<!-- gh-comment-id:2933235385 --> @yoin528 commented on GitHub (Jun 3, 2025): > Hi 已更新以包括思考 - 是否可以使用 and 重新下载(应该几乎是即时的)`qwen3``ollama pull qwen3:4b``ollama pull qwen3:8b` Thank you, it will be fine after updating the model qwen3:4b and qwen3:8b
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#32975