github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-07-16 05:05:37 -05:00
Code Issues 16.9k Packages Projects Releases 122 Wiki Activity

[GH-ISSUE #12159] gpt-oss models don't support parallel tool calling #85773

New Issue
Closed
opened 2026-05-10 01:35:11 -05:00 by GiteaMirror · 1 comment
GiteaMirror commented 2026-05-10 01:35:11 -05:00
Owner
Copy Link

Originally created by @pamelafox on GitHub (Sep 2, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/12159

Originally assigned to: @ParthSareen on GitHub.

What is the issue?

I tried a basic parallel tool calling example, where hosted OpenAI models like gpt-5 will happily issue multiple tool calls in the same response, but gpt-oss only ever calls a single tool. I imagine this is by design, but couldn't find documentation anywhere as to whether parallel tool calling is supported. Maybe it could be noted on https://github.com/ollama/ollama/blob/main/docs/openai.md (Maybe add under parameters "parallel_tool_calls - Not supported, always false")

Here's the example code:

import os

import openai
from dotenv import load_dotenv

# Setup the OpenAI client to use either Azure, OpenAI.com, or Ollama API
load_dotenv(override=True)
client = openai.OpenAI(base_url=os.environ["OLLAMA_ENDPOINT"], api_key="nokeyneeded")
MODEL_NAME = "gpt-oss:20b"

tools = [
    {
        "type": "function",
        "function": {
            "name": "lookup_weather",
            "description": "Lookup the weather for a given city name or zip code.",
            "parameters": {
                "type": "object",
                "properties": {
                    "city_name": {
                        "type": "string",
                        "description": "The city name",
                    },
                    "zip_code": {
                        "type": "string",
                        "description": "The zip code",
                    },
                },
                "additionalProperties": False,
            },
        },
    },
    {
        "type": "function",
        "function": {
            "name": "lookup_movies",
            "description": "Lookup movies playing in a given city name or zip code.",
            "parameters": {
                "type": "object",
                "properties": {
                    "city_name": {
                        "type": "string",
                        "description": "The city name",
                    },
                    "zip_code": {
                        "type": "string",
                        "description": "The zip code",
                    },
                },
                "additionalProperties": False,
            },
        },
    },
]

response = client.chat.completions.create(
    model=MODEL_NAME,
    messages=[
        {"role": "system", "content": "You are a tourism chatbot."},
        {"role": "user", "content": "is it rainy enough in sydney to watch movies and which ones are on?"},
    ],
    tools=tools,
    tool_choice="auto",
    reasoning_effort="medium"
)

print("Thinking...")
print (response.choices[0].message.reasoning)
print("Done thinking.")
print(f"Response from {MODEL_NAME} on {API_HOST}: \n")
for message in response.choices[0].message.tool_calls:
    # Potential bug: It did not output two tool calls- no support for parallel tool calling?
    print(message.function.name)
    print(message.function.arguments)

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @pamelafox on GitHub (Sep 2, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/12159 Originally assigned to: @ParthSareen on GitHub. ### What is the issue? I tried a basic parallel tool calling example, where hosted OpenAI models like gpt-5 will happily issue multiple tool calls in the same response, but gpt-oss only ever calls a single tool. I imagine this is by design, but couldn't find documentation anywhere as to whether parallel tool calling is supported. Maybe it could be noted on https://github.com/ollama/ollama/blob/main/docs/openai.md (Maybe add under parameters "parallel_tool_calls - Not supported, always false") Here's the example code: ``` import os import openai from dotenv import load_dotenv # Setup the OpenAI client to use either Azure, OpenAI.com, or Ollama API load_dotenv(override=True) client = openai.OpenAI(base_url=os.environ["OLLAMA_ENDPOINT"], api_key="nokeyneeded") MODEL_NAME = "gpt-oss:20b" tools = [ { "type": "function", "function": { "name": "lookup_weather", "description": "Lookup the weather for a given city name or zip code.", "parameters": { "type": "object", "properties": { "city_name": { "type": "string", "description": "The city name", }, "zip_code": { "type": "string", "description": "The zip code", }, }, "additionalProperties": False, }, }, }, { "type": "function", "function": { "name": "lookup_movies", "description": "Lookup movies playing in a given city name or zip code.", "parameters": { "type": "object", "properties": { "city_name": { "type": "string", "description": "The city name", }, "zip_code": { "type": "string", "description": "The zip code", }, }, "additionalProperties": False, }, }, }, ] response = client.chat.completions.create( model=MODEL_NAME, messages=[ {"role": "system", "content": "You are a tourism chatbot."}, {"role": "user", "content": "is it rainy enough in sydney to watch movies and which ones are on?"}, ], tools=tools, tool_choice="auto", reasoning_effort="medium" ) print("Thinking...") print (response.choices[0].message.reasoning) print("Done thinking.") print(f"Response from {MODEL_NAME} on {API_HOST}: \n") for message in response.choices[0].message.tool_calls: # Potential bug: It did not output two tool calls- no support for parallel tool calling? print(message.function.name) print(message.function.arguments) ``` ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-05-10 01:35:11 -05:00
GiteaMirror commented 2026-05-10 01:35:13 -05:00
Author
Owner
Copy Link

@ParthSareen commented on GitHub (Sep 2, 2025):

Hi @pamelafox - gpt-oss does not support parallel tool calling. Don't think we have it in the docs but will add! You can use the following as reference: https://github.com/ollama/ollama-python/blob/main/examples/gpt-oss-tools-stream.py

<!-- gh-comment-id:3247053451 --> @ParthSareen commented on GitHub (Sep 2, 2025): Hi @pamelafox - gpt-oss does not support parallel tool calling. Don't think we have it in the docs but will add! You can use the following as reference: https://github.com/ollama/ollama-python/blob/main/examples/gpt-oss-tools-stream.py
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#85773
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?