langchain - 💡(How to fix) Fix ChatAnthropic.bind_tools mutates caller-provided tool_choice dict

Submission checklist

• This is a bug, not a usage question.
• I added a clear and descriptive title that summarizes this issue.
• 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).
• This is not related to the langchain-community package.
• I posted a self-contained, minimal, reproducible example. A maintainer can copy it and run it AS IS.

Package (Required)

• langchain
• langchain-openai
• langchain-anthropic
• langchain-classic
• langchain-core
• langchain-model-profiles
• langchain-tests
• langchain-text-splitters
• langchain-chroma
• langchain-deepseek
• langchain-exa
• langchain-fireworks
• langchain-groq
• langchain-huggingface
• langchain-mistralai
• langchain-nomic
• langchain-ollama
• langchain-openrouter
• langchain-perplexity
• langchain-qdrant
• langchain-xai
• Other / not sure / general

Related Issues / PRs

I did not find an existing open issue or PR specifically about ChatAnthropic.bind_tools mutating a caller-provided tool_choice dict.

Reproduction Steps / Example Code (Python)

from langchain_anthropic import ChatAnthropic

tool_choice = {"type": "tool", "name": "GetWeather"}

model = ChatAnthropic(model="claude-sonnet-4-5-20250929")

model.bind_tools(
    tools=[
        {
            "name": "GetWeather",
            "description": "Get weather",
            "input_schema": {"type": "object", "properties": {}},
        }
    ],
    tool_choice=tool_choice,
    parallel_tool_calls=False,
)

assert tool_choice == {"type": "tool", "name": "GetWeather"}

Error Message and Stack Trace (if applicable)

No exception is raised. The issue is that tool_choice is mutated in place:

{
    "type": "tool",
    "name": "GetWeather",
    "disable_parallel_tool_use": True,
}

Description

When ChatAnthropic.bind_tools receives tool_choice as a dict and parallel_tool_calls is set, it stores the caller-provided dict directly in kwargs["tool_choice"], then adds disable_parallel_tool_use to it.

This mutates the original tool_choice object passed by the caller. If the same dict is reused across calls, the disable_parallel_tool_use option can leak into later calls unexpectedly.

A minimal fix would be to copy the tool_choice dict before adding disable_parallel_tool_use, preserving the existing Anthropic wire payload while avoiding mutation of caller-owned input. I’d also add a focused unit test covering the no-mutation behavior.

Would this be an acceptable scoped fix? If so, could you please assign this to me?

System Info

System Information

OS: Linux OS Version: #1 SMP PREEMPT_DYNAMIC Mon Dec 1 20:46:23 UTC 2025 Python Version: 3.10.12 (main, Mar 3 2026, 11:56:32) [GCC 11.4.0]

Package Information

langchain_core: 1.3.3 langsmith: 0.8.1 langchain_protocol: 0.0.15 langgraph_sdk: 0.3.14

Optional packages not installed

deepagents deepagents-cli

Other Dependencies

httpx: 0.28.1 jsonpatch: 1.33 orjson: 3.11.8 packaging: 24.2 pydantic: 2.12.5 pytest: 8.4.2 pyyaml: 5.4.1 requests: 2.32.5 requests-toolbelt: 1.0.0 tenacity: 9.1.2 typing-extensions: 4.15.0 uuid-utils: 0.14.1 websockets: 16.0 wrapt: 1.17.2 xxhash: 3.5.0 zstandard: 0.25.0