github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-06 16:11:34 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #12907] "/set nothink" not disabling reasoning output in Qwen3:4B #8555

New Issue
Closed
opened 2026-04-12 21:16:21 -05:00 by GiteaMirror · 2 comments
GiteaMirror commented 2026-04-12 21:16:21 -05:00
Owner
Copy Link

Originally created by @Saleh7127 on GitHub (Nov 2, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12907

What is the issue?

When running the Qwen3:4B model, the model still outputs its internal reasoning (“thinking”) even when the thinking mode is explicitly disabled.

I have tried both methods:

  • Setting "think": false in the /api/chat JSON payload
  • Using the CLI command "/set nothink" before the prompt

Neither disables the reasoning traces; the model still prints intermediate reasoning steps before the final answer.

Image Image

Relevant log output

OS

macOS

GPU

Apple

CPU

Apple

Ollama version

0.12.9

Originally created by @Saleh7127 on GitHub (Nov 2, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12907 ### What is the issue? When running the Qwen3:4B model, the model still outputs its internal reasoning (“thinking”) even when the thinking mode is explicitly disabled. I have tried both methods: - Setting "think": false in the /api/chat JSON payload - Using the CLI command "/set nothink" before the prompt Neither disables the reasoning traces; the model still prints intermediate reasoning steps before the final answer. <img width="546" height="307" alt="Image" src="https://github.com/user-attachments/assets/346f5e7e-9ab4-437b-ab4f-78e0b5b95947" /> <img width="852" height="397" alt="Image" src="https://github.com/user-attachments/assets/f9af4d68-642e-4d3c-8e20-bd2970d6496b" /> ### Relevant log output ```shell ``` ### OS macOS ### GPU Apple ### CPU Apple ### Ollama version 0.12.9
GiteaMirror added the bug label 2026-04-12 21:16:21 -05:00
GiteaMirror commented 2026-04-12 21:16:23 -05:00
Author
Owner
Copy Link

@ronething commented on GitHub (Nov 2, 2025):

i found qwen3-vl:8b also had this issue.

Image
<!-- gh-comment-id:3478103492 --> @ronething commented on GitHub (Nov 2, 2025): i found `qwen3-vl:8b` also had this issue. <img width="1216" height="322" alt="Image" src="https://github.com/user-attachments/assets/576c1445-debd-42d3-ad8a-d8fb75663bdd" />
GiteaMirror commented 2026-04-12 21:16:23 -05:00
Author
Owner
Copy Link

@jmorganca commented on GitHub (Nov 2, 2025):

Hi all, similar to qwen3, qwen3-vl has different versions for whether the model can think or not. For the non-thinking version, use ollama run qwen3-vl:8b-instruct

<!-- gh-comment-id:3478500546 --> @jmorganca commented on GitHub (Nov 2, 2025): Hi all, similar to `qwen3`, `qwen3-vl` has different versions for whether the model can think or not. For the non-thinking version, use `ollama run qwen3-vl:8b-instruct`
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#8555
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?