feat(core): (v1) restore separate type for AIMessage.tool_calls

[ccurme]

"type" needs to be a required key for content blocks, including tool calls in content blocks, for type discrimination.

This has not a required key for the tool calls in AIMessage.tool_calls. If we require that content ToolCall and .tool_calls ToolCall both require "type", we introduce type errors in user code:

msg.tool_calls = [{"name": "foo", "args": {"a": "b"}, "id": "abc-123"}]

Any pydantic classes expecting tool calls would also raise validation errors if "type" is not provided— I realized this issue when this happened to some test classes in langgraph.

For now, we restore the old type, so that for both AIMessage.tool_calls and AIMessage.tool_call_chunks we have separate types that do not require the "type" key.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Preview	Comments	Updated (UTC)
langchain	Ignored	Preview		Aug 25, 2025 1:18pm

[codspeed-hq]

CodSpeed WallTime Performance Report

Merging #32668 will degrade performances by 28.39%

_{Comparing cc/1.0/standard_content_tc_type (6fecdf1) with cc/1.0/standard_content (d111965)¹}

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

Summary

❌ 1 regressions
✅ 12 untouched benchmarks

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
❌	test_import_time[HumanMessage]	211.8 ms	295.8 ms	-28.39%

Footnotes

1. No successful run was found on cc/1.0/standard_content (26833f2) during the generation of this report, so 823c15c was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #32668 will degrade performances by 60.72%

_{Comparing cc/1.0/standard_content_tc_type (6fecdf1) with cc/1.0/standard_content (d111965)}

Summary

⚡ 1 improvements
❌ 1 (👁 1) regressions
✅ 12 untouched benchmarks

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
👁	test_stream_time	15.9 ms	40.5 ms	-60.72%
⚡	test_stream_time	1,209.3 ms	942.3 ms	+28.35%

[mdrxy]

ewwwwwwwwwwwww

[mdrxy]

assignment doesn't add a type key right?