LangGraph 1.2.15 fixes DeltaChannel and checkpoint bugs
Updated LangGraph
Summary
LangGraph 1.2.15, published October 12, 2026, is a fix release that tightens DeltaChannel, checkpoint, interrupt and ToolNode resume handling.
- Tool
- LangGraph
- Category
- Frameworks
- Date
- Open source
- Yes
- License
- MIT
- Pricing
- Not stated in the sources we cited.
LangGraph 1.2.15 was published on October 12, 2026 in the langchain-ai/langgraph repository on GitHub. According to the release notes, it is a set of fixes since 1.2.14, mostly around DeltaChannel handling, checkpoints and interrupts.
Release facts
| Fact | Detail |
|---|---|
| Product | LangGraph, described in the docs as an orchestration runtime |
| Version | langgraph 1.2.15 (tag 1.2.15) |
| Release date | October 12, 2026 |
| License | MIT, as reported by the GitHub API |
| Where to get it | GitHub release 1.2.15 |
What changed
The release notes list fixes rather than new features. The ones that matter most for people running stateful agents:
- Interrupts: a run now stops at an
interrupt_beforebreakpoint when a drain is requested before it. - Resume values: ToolNode no longer swallows invalid resume values.
- Streaming:
astream_eventsno longer drops control and interrupts on v1/v2. - Checkpoints: a checkpoint is no longer saved past a failed DeltaChannel write, and
update_stateinput writes are stored on the checkpoint they update. - Messages: every message written to a messages DeltaChannel now gets an id.
- Replay: delta channel replay is ordered by task path, and a resumed exit-mode run's delta writes replay in live order.
The release also bumps the notebook dependency from 7.5.7 to 7.6.3 in /libs/langgraph.
Questions
Is this a feature release? The release notes list only fixes, a dependency bump and the release commit under "Changes since 1.2.14". They do not describe new features.
Is LangGraph open source? The GitHub repository reports an MIT license. The cited sources do not state hosted pricing.
What is LangGraph for? The LangChain docs describe it as the orchestration runtime offering durable execution, streaming, human-in-the-loop and persistence.
See the LangGraph tool hub, the frameworks category and the related Mastra core 1.76.0 launch for more.
Alternatives
| Tool | License | Pricing note |
|---|---|---|
LangGraphThis launchLangGraph is an open-source orchestration runtime for building stateful, resilient agents. | MIT | Not stated |
MastraTypeScript agent framework. Mastra Code is the terminal agent; AgentController is the extracted loop, in @mastra/core 1.47.0 or later. | Open source | Mastra Code and AgentController are announced as open source. No license fee is stated in the posts we cited. |
adk-pythonAn open-source, code-first Python toolkit for building, evaluating and deploying AI agents. | Apache-2.0 | Not stated |
AgnoAgno is an orchestration and service layer for agents, exposing them through a REST API and an MCP server. | Apache-2.0 | Not stated |
OpenHandsMIT-licensed project to build and run AI coding agents: start locally, automate workflows, use any model. | MIT | The OpenHands repository is MIT-licensed. Hosted plans are not priced on this page. |
Questions
What is LangGraph?
The LangChain docs describe LangGraph as the orchestration runtime with durable execution, streaming, human-in-the-loop and persistence. The repository description is "Build resilient agents."
Is LangGraph open source?
The GitHub API reports an MIT license for the langchain-ai/langgraph repository.
What is the latest launch covered here?
LangGraph 1.2.15, published October 12, 2026, which the release notes list as a set of fixes.
Filed on Agent Radar. Claims in this note are limited to the sources above.