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 #12180] Can't disable thinking on deepseek-r1 with openai sdk #33860

New Issue
Closed
opened 2026-04-22 16:58:25 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-04-22 16:58:25 -05:00
Owner
Copy Link

Originally created by @gvsolent on GitHub (Sep 4, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12180

Hello,

I need to disable reasoning on deepseek-r1.

I'd like to be able to do this using openai sdk. This would help to call various LLMs/providers in the same way.

I was able to disable reasoning using the post method, but not with open ai sdk.

Note that I use different endpoints for each. i don't know if that's the problem.

Could you help with that?

post methods :

import requests
import json

# URL de ton instance Ollama
OLLAMA_URL = "myurl/api/chat"

# Modèle (ex: deepseek-r1, qwen2.5:7b, etc.) deepseek-r1:70b
MODEL = "deepseek-r1:70b"

# Prompt simple
PROMPT = "how many r in the word strawberry?"

# Payload minimal avec paramètre "think"
payload = {
    "model": MODEL,
    "messages": [
        {"role": "user", "content": PROMPT}
    ],
    "think": False, 
    "stream": False
}

print("➡️ Payload envoyé :")
print(json.dumps(payload, indent=2))

# Appel HTTP POST
resp = requests.post(OLLAMA_URL, json=payload)

print("\n➡️ Status:", resp.status_code)
print("\n➡️ Réponse brute :")
print(resp.text)

try:
    data = resp.json()
    print("\n➡️ Contenu extrait :")
    print(json.dumps(data, indent=2))
except Exception as e:
    print("Erreur parsing JSON:", e)

With openai sdk

from openai import OpenAI
PROMPT = "how many r in the word strawberry?"
client = OpenAI(
    api_key="sk-xxxxxx",
    base_url="myurl/v1"
)

completion = client.chat.completions.create(
    model="deepseek-r1:70b",
    messages=[{"role": "user", "content": PROMPT}],
    temperature=0,
    max_tokens=2000,
    extra_body={"think": False}   # 👈 désactive le raisonnement
)

print(completion.choices[0].message)
Originally created by @gvsolent on GitHub (Sep 4, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12180 Hello, I need to disable reasoning on deepseek-r1. I'd like to be able to do this using openai sdk. This would help to call various LLMs/providers in the same way. I was able to disable reasoning using the post method, but not with open ai sdk. Note that I use different endpoints for each. i don't know if that's the problem. Could you help with that? post methods : ``` import requests import json # URL de ton instance Ollama OLLAMA_URL = "myurl/api/chat" # Modèle (ex: deepseek-r1, qwen2.5:7b, etc.) deepseek-r1:70b MODEL = "deepseek-r1:70b" # Prompt simple PROMPT = "how many r in the word strawberry?" # Payload minimal avec paramètre "think" payload = { "model": MODEL, "messages": [ {"role": "user", "content": PROMPT} ], "think": False, "stream": False } print("➡️ Payload envoyé :") print(json.dumps(payload, indent=2)) # Appel HTTP POST resp = requests.post(OLLAMA_URL, json=payload) print("\n➡️ Status:", resp.status_code) print("\n➡️ Réponse brute :") print(resp.text) try: data = resp.json() print("\n➡️ Contenu extrait :") print(json.dumps(data, indent=2)) except Exception as e: print("Erreur parsing JSON:", e) ``` With openai sdk ``` from openai import OpenAI PROMPT = "how many r in the word strawberry?" client = OpenAI( api_key="sk-xxxxxx", base_url="myurl/v1" ) completion = client.chat.completions.create( model="deepseek-r1:70b", messages=[{"role": "user", "content": PROMPT}], temperature=0, max_tokens=2000, extra_body={"think": False} # 👈 désactive le raisonnement ) print(completion.choices[0].message) ```
GiteaMirror added the feature request label 2026-04-22 16:58:25 -05:00
GiteaMirror commented 2026-04-22 16:58:27 -05:00
Author
Owner
Copy Link

@rick-github commented on GitHub (Sep 4, 2025):

#11012

<!-- gh-comment-id:3254057136 --> @rick-github commented on GitHub (Sep 4, 2025): #11012
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#33860
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?