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 #12820] qwen3-vl:32b /set nothink not work and doesn't support parallel requests #70554

New Issue
Closed
opened 2026-05-04 21:57:13 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-04 21:57:13 -05:00
Owner
Copy Link

Originally created by @chuanyu1 on GitHub (Oct 29, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12820

What is the issue?

  1. Unable to switch on or off the thinking mode
    ~/wwork/ollama_w$ ollama -v
    ollama version is 0.12.7-rc0
>>> hi /no_think
Thinking...
Okay, the user is sending "hi /no_think" again. Let me look back at the history to      
understand what's going on. 

First, the user started with "hi", then "set", then "hi" again. The assistant           
asked for clarification about "set". Then the user sent "hi /nothink" and "hi           
/no_think". The assistant responded^C


>>> /set nothink
Set 'nothink' mode.
>>> hi
Thinking...
Okay, the user just said "hi" again after a previous interaction where they first       
said "hi", then "set", and now back to "hi". Hmm, this feels like a pattern. 

Let me recap: 
First message was a simple "hi" - I responded warmly with an emoji and offered          
help. 
Then they sent "set" - I asked for clarification since it was ambiguous. 
Now they're back to "hi" - probably testing if I recognize the pattern or just          
being polite. ```
2. parallel requests can't be supported,Other models （eg qwen2.5:32b）don't have this problem


### Relevant log output

```shell

OS

Linux

GPU

Nvidia

CPU

Intel

Ollama version

0.12.7-rc0

Originally created by @chuanyu1 on GitHub (Oct 29, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12820 ### What is the issue? 1. Unable to switch on or off the thinking mode ~/wwork/ollama_w$ ollama -v ollama version is 0.12.7-rc0 ``` >>> hi /no_think Thinking... Okay, the user is sending "hi /no_think" again. Let me look back at the history to understand what's going on. First, the user started with "hi", then "set", then "hi" again. The assistant asked for clarification about "set". Then the user sent "hi /nothink" and "hi /no_think". The assistant responded^C >>> /set nothink Set 'nothink' mode. >>> hi Thinking... Okay, the user just said "hi" again after a previous interaction where they first said "hi", then "set", and now back to "hi". Hmm, this feels like a pattern. Let me recap: First message was a simple "hi" - I responded warmly with an emoji and offered help. Then they sent "set" - I asked for clarification since it was ambiguous. Now they're back to "hi" - probably testing if I recognize the pattern or just being polite. ``` 2. parallel requests can't be supported,Other models （eg qwen2.5:32b）don't have this problem ### Relevant log output ```shell ``` ### OS Linux ### GPU Nvidia ### CPU Intel ### Ollama version 0.12.7-rc0
GiteaMirror added the bug label 2026-05-04 21:57:13 -05:00
GiteaMirror commented 2026-05-04 21:57:16 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Oct 29, 2025):

qwen3-vl comes in thinking and non-thinking variants. If you want non-thinking, use the instruct variant.

Set OLLAMA_NUM_PARALLEL to increase parallelism.

<!-- gh-comment-id:3460569201 --> @rick-github commented on GitHub (Oct 29, 2025): qwen3-vl comes in thinking and non-thinking variants. If you want non-thinking, use the [instruct](https://ollama.com/library/qwen3-vl:32b-instruct) variant. Set [`OLLAMA_NUM_PARALLEL`](https://docs.ollama.com/faq#how-does-ollama-handle-concurrent-requests%3F) to increase parallelism.
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#70554
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?