thinking_block field returned by Claude model ignored by ChatOpenAI().invoke()

[vigneshvs]

Checked other resources

• I added a very descriptive title to this issue.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.
• I am sure that this is a bug in LangChain rather than my code.
• The bug is not resolved by updating to the latest stable version of LangChain (or the specific integration package).

Example Code

import os
os.environ["OPENAI_BASE_URL"] = <locally_hosted_API_service_that_calls_bedrock_or_openai>
os.environ["OPENAI_API_KEY"] = <auth_key_for_service>

from langchain_openai import ChatOpenAI

llm = ChatOpenAI(
        temperature=1,
        max_tokens=20000,
        model="us.anthropic.claude-3-7-sonnet-20250219-v1:0", 
        extra_body={"thinking": {"type": "enabled","budget_tokens": 16000}}
)
messages = [
    ("user", "hello!"),
]
ai_msg = llm.invoke(messages)
print(ai_msg)

Error Message and Stack Trace (if applicable)

Response

content="Hello there! It's nice to meet you. I'm an AI assistant ready to help with information, answer questions, or just chat. How are you today, and is there something specific I can help you with?" additional_kwargs={'refusal': None} response_metadata={'token_usage': {'completion_tokens': 93, 'prompt_tokens': 37, 'total_tokens': 130, 'completion_tokens_details': None, 'prompt_tokens_details': {'audio_tokens': None, 'cached_tokens': 0}, 'cache_creation_input_tokens': 0, 'cache_read_input_tokens': 0}, 'model_name': 'us.anthropic.claude-3-7-sonnet-20250219-v1:0', 'system_fingerprint': None, 'id': '<some_uuid>', 'finish_reason': 'stop', 'logprobs': None} id='<some_uuid>' usage_metadata={'input_tokens': 37, 'output_tokens': 93, 'total_tokens': 130, 'input_token_details': {'cache_read': 0}, 'output_token_details': {}}

The following reasoning_content and thinking_blocks fields are not found in the response obtained via langchain -

"choices":[{"index":0,"message":{"content":"Hi there! Welcome! I'm an AI assistant and I'm happy to help with any questions, provide information, or just chat. What can I do for you today?","role":"assistant","tool_calls":null,"function_call":null,"reasoning_content":"Hello! This is a simple greeting from the user. I should respond in a friendly, welcoming manner that acknowledges their greeting and shows I'm ready to assist them with whatever they need.","thinking_blocks":[{"type":"thinking","thinking":"Hello! This is a simple greeting from the user. I should respond in a friendly, welcoming manner that acknowledges their greeting and shows I'm ready to assist them with whatever they need.","signature":"ErcBCkgIARABGAIiQBqFPuLSq3mX010ufxwlrJY5kdiRQwQF1VZ7izukmsv0ARO43wE/knHBrcL9YmBB4HPu4a9j6gVlcwINVAoFHccSDFV1vU70AIK3wRo00BoMRfVPi41wJw6xJNLAIjA3jG6QLT38KXp7Wp/lI8zJNUGhQkqEI8/eonaq6UMDuGMsnbsSjVyrv6iaiMljnAoqHfcMBcGK+c8ZGgA7gL6P1pyneTDBjh+BORWw2fRu"}]},"finish_reason":"stop"}]

Note: Response variation observed due to separate LLM calls

Description

We use ChatOpenAI() as the interface for interacting with a locally API service that is based on LiteLLM and calls OpenAI or Bedrock depending on the model name. With Anthropic Claude sonnet 3.7's extended thinking, the model provides reasoning_content and thinking_blocks in the response fields. However, this is swallowed by langchain and returns only the vanilla fields.

System Info

$ python -m langchain_core.sys_info



System Information
------------------
> OS:  Darwin
> OS Version:  Darwin Kernel Version 23.6.0: Thu Dec 19 20:44:43 PST 2024; root:xnu-10063.141.1.703.2~1/RELEASE_ARM64_T6020
> Python Version:  3.9.13 (main, Sep 27 2023, 15:09:55) 
[Clang 15.0.0 (clang-1500.0.40.1)]

Package Information
-------------------
> langchain_core: 0.3.48
> langchain: 0.3.21
> langsmith: 0.3.18
> langchain_aws: 0.2.17
> langchain_openai: 0.3.10
> langchain_text_splitters: 0.3.7

Optional packages not installed
-------------------------------
> langserve

Other Dependencies
------------------
> async-timeout<5.0.0,>=4.0.0;: Installed. No version info available.
> boto3: 1.37.21
> httpx: 0.28.1
> jsonpatch<2.0,>=1.33: Installed. No version info available.
> langchain-anthropic;: Installed. No version info available.
> langchain-aws;: Installed. No version info available.
> langchain-azure-ai;: Installed. No version info available.
> langchain-cohere;: Installed. No version info available.
> langchain-community;: Installed. No version info available.
> langchain-core<1.0.0,>=0.3.45: Installed. No version info available.
> langchain-core<1.0.0,>=0.3.48: Installed. No version info available.
> langchain-deepseek;: Installed. No version info available.
> langchain-fireworks;: Installed. No version info available.
> langchain-google-genai;: Installed. No version info available.
> langchain-google-vertexai;: Installed. No version info available.
> langchain-groq;: Installed. No version info available.
> langchain-huggingface;: Installed. No version info available.
> langchain-mistralai;: Installed. No version info available.
> langchain-ollama;: Installed. No version info available.
> langchain-openai;: Installed. No version info available.
> langchain-text-splitters<1.0.0,>=0.3.7: Installed. No version info available.
> langchain-together;: Installed. No version info available.
> langchain-xai;: Installed. No version info available.
> langsmith-pyo3: Installed. No version info available.
> langsmith<0.4,>=0.1.125: Installed. No version info available.
> langsmith<0.4,>=0.1.17: Installed. No version info available.
> numpy: 1.26.4
> openai-agents: Installed. No version info available.
> openai<2.0.0,>=1.68.2: Installed. No version info available.
> opentelemetry-api: Installed. No version info available.
> opentelemetry-exporter-otlp-proto-http: Installed. No version info available.
> opentelemetry-sdk: Installed. No version info available.
> orjson: 3.10.16
> packaging: 24.2
> packaging<25,>=23.2: Installed. No version info available.
> pydantic: 2.10.6
> pydantic<3.0.0,>=2.5.2;: Installed. No version info available.
> pydantic<3.0.0,>=2.7.4: Installed. No version info available.
> pydantic<3.0.0,>=2.7.4;: Installed. No version info available.
> pytest: Installed. No version info available.
> PyYAML>=5.3: Installed. No version info available.
> requests: 2.32.3
> requests-toolbelt: 1.0.0
> requests<3,>=2: Installed. No version info available.
> rich: Installed. No version info available.
> SQLAlchemy<3,>=1.4: Installed. No version info available.
> tenacity!=8.4.0,<10.0.0,>=8.1.0: Installed. No version info available.
> tiktoken<1,>=0.7: Installed. No version info available.
> typing-extensions>=4.7: Installed. No version info available.
> zstandard: 0.23.0

[andrasfe]

you should use ChatAnthropic instead of the OpenAI equivalent. The following works for me:

import os
from dotenv import load_dotenv

load_dotenv()

from langchain_anthropic import ChatAnthropic

llm = ChatAnthropic(
    model="claude-3-7-sonnet-latest", 
    max_tokens=5000,
    thinking={"type": "enabled", "budget_tokens": 2000}
)
messages = [
    ("user", "hello!"),
]
ai_msg = llm.invoke(messages)
print(ai_msg)