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 #10473] Add support for enable_thinking=True #6887

New Issue
Closed
opened 2026-04-12 18:45:12 -05:00 by GiteaMirror · 9 comments
GiteaMirror commented 2026-04-12 18:45:12 -05:00
Owner
Copy Link

Originally created by @lknik on GitHub (Apr 29, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10473

In Qwen3, enable_thinking=True disables the part. It would be useful to allow its use in programmatic calls, without the need to include it in the prompt.

Specifically, I would suggest:

  1. Having a command line switch
  2. CLI command
  3. Make the examples below working:
import openai

client = openai.OpenAI( ... )

response = client.chat.completions.create(
    model="qwen3",
    messages=[{"role": "user", "content": "Hello"}],
    extra_body={"options": {"enable_thinking": False}}
)

import ollama

response = ollama.chat(
    model='qwen3',
    messages=[{'role': 'user', 'content': 'Hello'}],
    options={'enable_thinking': False}
)
Originally created by @lknik on GitHub (Apr 29, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10473 In Qwen3, enable_thinking=True disables the <think> part. It would be useful to allow its use in programmatic calls, without the need to include it in the prompt. Specifically, I would suggest: 1) Having a command line switch 2) CLI command 3) Make the examples below working: ``` import openai client = openai.OpenAI( ... ) response = client.chat.completions.create( model="qwen3", messages=[{"role": "user", "content": "Hello"}], extra_body={"options": {"enable_thinking": False}} ) ``` ``` import ollama response = ollama.chat( model='qwen3', messages=[{'role': 'user', 'content': 'Hello'}], options={'enable_thinking': False} ) ```
GiteaMirror added the feature request label 2026-04-12 18:45:12 -05:00
GiteaMirror commented 2026-04-12 18:45:14 -05:00
Author
Owner
Copy Link

@hchasens commented on GitHub (Apr 29, 2025):

proper API support would be great.

<!-- gh-comment-id:2839148083 --> @hchasens commented on GitHub (Apr 29, 2025): proper API support would be great.
GiteaMirror commented 2026-04-12 18:45:14 -05:00
Author
Owner
Copy Link

@defaultsecurity commented on GitHub (May 12, 2025):

Yes, please allow this option.

<!-- gh-comment-id:2873934058 --> @defaultsecurity commented on GitHub (May 12, 2025): Yes, please allow this option.
GiteaMirror commented 2026-04-12 18:45:15 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (May 30, 2025):

https://github.com/ollama/ollama/releases/tag/v0.9.0

<!-- gh-comment-id:2921919466 --> @rick-github commented on GitHub (May 30, 2025): https://github.com/ollama/ollama/releases/tag/v0.9.0
GiteaMirror commented 2026-04-12 18:45:16 -05:00
Author
Owner
Copy Link

@lknik commented on GitHub (Jun 2, 2025):

@rick-github

Thanks for the update. However:

I just tried:

$ ollama -v
ollama version is 0.9.0

$ curl http://localhost:11434/api/chat -d '{ "model": "qwen3:30b-a3b-q8_0", "messages": [ { "role": "user", "content": "Why is the sky blue?" } ], "think": true }'
{"error":"registry.ollama.ai/library/qwen3:30b-a3b-q8_0 does not support thinking"}

$ ollama run qwen3:30b-a3b-q8_0

a

Okay, the user just sent "a". That's it. No other text. Hmm, maybe they're testing the waters or just made a typo. I
should respond politely and ask how I can assist them. Let me make sure to keep it friendly and open-ended. Maybe say
something like, "Hello! How can I assist you today?" That should encourage them to provide more information if they
need help with something specific.

Hello! How can I assist you today? 😊

<!-- gh-comment-id:2930238085 --> @lknik commented on GitHub (Jun 2, 2025): @rick-github Thanks for the update. However: I just tried: $ ollama -v ollama version is 0.9.0 $ curl http://localhost:11434/api/chat -d '{ "model": "qwen3:30b-a3b-q8_0", "messages": [ { "role": "user", "content": "Why is the sky blue?" } ], "think": true }' {"error":"registry.ollama.ai/library/qwen3:30b-a3b-q8_0 does not support thinking"} $ ollama run qwen3:30b-a3b-q8_0 >>> a <think> Okay, the user just sent "a". That's it. No other text. Hmm, maybe they're testing the waters or just made a typo. I should respond politely and ask how I can assist them. Let me make sure to keep it friendly and open-ended. Maybe say something like, "Hello! How can I assist you today?" That should encourage them to provide more information if they need help with something specific. </think> Hello! How can I assist you today? 😊
GiteaMirror commented 2026-04-12 18:45:17 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 2, 2025):

Re-pull the model.

<!-- gh-comment-id:2930284835 --> @rick-github commented on GitHub (Jun 2, 2025): Re-pull the model.
GiteaMirror commented 2026-04-12 18:45:18 -05:00
Author
Owner
Copy Link

@lknik commented on GitHub (Jun 2, 2025):

Ahh, right. Would the flag work for non-ollama models though?

<!-- gh-comment-id:2930315481 --> @lknik commented on GitHub (Jun 2, 2025): Ahh, right. Would the flag work for non-ollama models though?
GiteaMirror commented 2026-04-12 18:45:19 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Jun 2, 2025):

No, but you can merge the template from an ollama model with a non-ollama model to make a thinking non-ollama model: https://github.com/ollama/ollama/issues/10914#issuecomment-2926168350

<!-- gh-comment-id:2930327096 --> @rick-github commented on GitHub (Jun 2, 2025): No, but you can merge the template from an ollama model with a non-ollama model to make a thinking non-ollama model: https://github.com/ollama/ollama/issues/10914#issuecomment-2926168350
GiteaMirror commented 2026-04-12 18:45:20 -05:00
Author
Owner
Copy Link

@lknik commented on GitHub (Jun 2, 2025):

Thanks! It won't work with Python's openai library?

<!-- gh-comment-id:2930893645 --> @lknik commented on GitHub (Jun 2, 2025): Thanks! It won't work with Python's openai library?
GiteaMirror commented 2026-04-12 18:45:21 -05:00
Author
Owner
Copy Link

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

OpenAI library can not worked with extra_body

<!-- gh-comment-id:2934614702 --> @elvizlai commented on GitHub (Jun 3, 2025): OpenAI library can not worked with extra_body
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 feature request
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#6887
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?