Issues with function calling with local Ollama

[mirix]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

from langchain_core.tools import tool
from langchain_core.messages import HumanMessage
from langchain_ollama import ChatOllama
from langchain_core.output_parsers import StrOutputParser
from langchain_core.prompts import ChatPromptTemplate

# Mock function

@tool
def book_appointment():
    """ Book appointment """
    book_slot = dparser.parse("monkey 2010-07-10 love banana", fuzzy=True)
    book_slot = datetime.strftime(book_slot, '%A %-d %B %Y %-I %p')
    print(book_slot)
    return book_slot

prompt = ChatPromptTemplate.from_messages([('system', System), ('user', '{question}')])

# Create chain

#prompt = ChatPromptTemplate.from_template(template)
model = ChatOllama(model='llama3.1:8b-instruct-q8_0', temperature=0)
model = model.bind_tools([book_appointment])
chain = prompt | model

result = chain.invoke({'context': context, 'avaliable_slots': slots, 'question': user_input})

Description

I am trying to use LangChain with local Ollama to execute function.

However, it produces no output contents:

ASSISTANT: content='' additional_kwargs={} response_metadata={'model': 'llama3.1:8b-instruct-q8_0', 'created_at': '2024-11-07T14:21:57.65997886Z', 'message': {'role': 'assistant', 'content': '', 'tool_calls': [{'function': {'name': 'book_appointment', 'arguments': {'date': 'Sunday 13 October 2024', 'time': '9 AM'}}}]}, 'done_reason': 'stop', 'done': True, 'total_duration': 629731509, 'load_duration': 10817049, 'prompt_eval_count': 597, 'prompt_eval_duration': 109600000, 'eval_count': 31, 'eval_duration': 376007000} id='run-ea7155b8-eaea-494a-b74d-c98c276557d5-0' tool_calls=[{'name': 'book_appointment', 'args': {'date': 'Sunday 13 October 2024', 'time': '9 AM'}, 'id': 'cabd3450-02a7-4a97-85e6-6438af56cdbc', 'type': 'tool_call'}] usage_metadata={'input_tokens': 597, 'output_tokens': 31, 'total_tokens': 628}

And it seems that is attempting to pass arguments to a function that needs none.

The function is not executed and there is no error message.

Using Llama on the cloud is not an option. Other local models could do.

System Info

pip freeze | grep langchain  127 ✘
langchain==0.3.7
langchain-community==0.3.5
langchain-core==0.3.15
langchain-experimental==0.3.3
langchain-ollama==0.2.0
langchain-text-splitters==0.3.2

Linux

Python 3.12.6

python -m langchain_core.sys_info

System Information

OS: Linux
OS Version: #1 SMP PREEMPT_DYNAMIC Tue Oct 8 03:24:49 UTC 2024
Python Version: 3.12.6 (main, Sep 8 2024, 13:18:56) [GCC 14.2.1 20240805]

Package Information

langchain_core: 0.3.15
langchain: 0.3.7
langchain_community: 0.3.5
langsmith: 0.1.140
langchain_experimental: 0.3.3
langchain_ollama: 0.2.0
langchain_text_splitters: 0.3.2

Optional packages not installed

langgraph
langserve

Other Dependencies

aiohttp: 3.10.10
async-timeout: 4.0.3
dataclasses-json: 0.6.7
httpx: 0.27.2
httpx-sse: 0.4.0
jsonpatch: 1.33
numpy: 1.26.4
ollama: 0.3.3
orjson: 3.10.10
packaging: 24.1
pydantic: 2.9.2
pydantic-settings: 2.6.0
PyYAML: 6.0.2
requests: 2.32.3
requests-toolbelt: 1.0.0
SQLAlchemy: 2.0.35
tenacity: 8.5.0
typing-extensions: 4.12.2