AI Agent News/stories/langgraph-sdk-0-4-14-adds-strict-resume-and-streammode-options.md

Frameworks

LangGraph SDK 0.4.14 adds strict_resume and StreamMode options

The langgraph-sdk Python package reached version 0.4.14 in the langchain-ai/langgraph repository. The update adds a strict_resume option to runs.create, stream and wait, and adds tools and lifecycle to StreamMode.

Source: github.com/langchain-ai/langgraph/releases/tag/sdk%3D%3D0.4.14

The release is a minor update to the LangGraph Python SDK, published on GitHub under the tag sdk==0.4.14. It lists three changes since sdk==0.4.13: the version bump itself, a strict_resume parameter, and new tools and lifecycle entries on StreamMode.

The strict_resume parameter is added to three SDK methods: runs.create, stream and wait. The changelog entry does not describe what strict_resume changes at runtime, so the behavior is not covered here.

The StreamMode change is listed under the same SDK and is titled as adding tools and lifecycle. The entry gives no further detail on the values or their output.

Why it matters if you build agents

If your agents call runs.create, stream or wait through the Python SDK, you can now pass strict_resume to those calls. Check the repository's code or docs before relying on it, since the release entry does not explain it.

If you use StreamMode to consume agent output, review the new tools and lifecycle options before changing your stream handling.

Get stories like this every weekday

One short email with the day's agent tooling news.

Subscribe