[PR #1720] [CLOSED] Added optional bool to disable parallel tool calling in prebuilt react agent #2257

Closed
opened 2026-02-20 17:46:39 -05:00 by yindo · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/langchain-ai/langgraph/pull/1720
Author: @akshatjain546
Created: 9/14/2024
Status: Closed

Base: mainHead: prebuilt-react-disable-parallel-tool


📝 Commits (1)

  • b8e2cb2 Added option to disable parallel tool calling

📊 Changes

1 file changed (+6 additions, -1 deletions)

View changed files

📝 libs/langgraph/langgraph/prebuilt/chat_agent_executor.py (+6 -1)

📄 Description

Hey folks,

I wanted to way to disable parallel tool calls in the prebuilt react agent so decided to modify this.
Reason: A lot of times the agent calls the same tool in parallel, leading to increased hallucinations and sub-optimal answers. I want the agent to call one tool at a time, think about the answer, and then call the next tool if necessary.
Unfortunately, this is not solved entirely through prompting.

Relevant Discussion: https://github.com/langchain-ai/langgraph/discussions/1461

cc: @vbarda


🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.

## 📋 Pull Request Information **Original PR:** https://github.com/langchain-ai/langgraph/pull/1720 **Author:** [@akshatjain546](https://github.com/akshatjain546) **Created:** 9/14/2024 **Status:** ❌ Closed **Base:** `main` ← **Head:** `prebuilt-react-disable-parallel-tool` --- ### 📝 Commits (1) - [`b8e2cb2`](https://github.com/langchain-ai/langgraph/commit/b8e2cb2ca4c8b4fd12f1c8446b2a56d845f7eb29) Added option to disable parallel tool calling ### 📊 Changes **1 file changed** (+6 additions, -1 deletions) <details> <summary>View changed files</summary> 📝 `libs/langgraph/langgraph/prebuilt/chat_agent_executor.py` (+6 -1) </details> ### 📄 Description Hey folks, I wanted to way to disable parallel tool calls in the prebuilt react agent so decided to modify this. Reason: A lot of times the agent calls the same tool in parallel, leading to increased hallucinations and sub-optimal answers. I want the agent to call one tool at a time, think about the answer, and then call the next tool if necessary. Unfortunately, this is not solved entirely through prompting. Relevant Discussion: https://github.com/langchain-ai/langgraph/discussions/1461 cc: @vbarda --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>
yindo added the pull-request label 2026-02-20 17:46:39 -05:00
yindo closed this issue 2026-02-20 17:46:40 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/langgraph#2257