[PR #1605] [MERGED] Reduce from 9s to 4s on benchmark graph #2174

Closed
opened 2026-02-20 17:46:27 -05:00 by yindo · 0 comments
Owner

📋 Pull Request Information

Original PR: https://github.com/langchain-ai/langgraph/pull/1605
Author: @nfcampos
Created: 9/4/2024
Status: Merged
Merged: 9/4/2024
Merged by: @nfcampos

Base: nc/3sep/perf-optimizationsHead: nc/4sep/perf-optimizations-ii


📝 Commits (1)

  • de21b4f Reduce from 9s to 4s on benchmark graph

📊 Changes

12 files changed (+446 additions, -148 deletions)

View changed files

📝 libs/langgraph/langgraph/graph/graph.py (+14 -8)
📝 libs/langgraph/langgraph/graph/state.py (+9 -16)
📝 libs/langgraph/langgraph/pregel/__init__.py (+7 -15)
📝 libs/langgraph/langgraph/pregel/algo.py (+2 -2)
📝 libs/langgraph/langgraph/pregel/read.py (+10 -8)
📝 libs/langgraph/langgraph/pregel/write.py (+38 -56)
📝 libs/langgraph/langgraph/utils/config.py (+3 -1)
📝 libs/langgraph/langgraph/utils/runnable.py (+301 -20)
📝 libs/langgraph/poetry.lock (+10 -5)
📝 libs/langgraph/tests/__snapshots__/test_pregel.ambr (+39 -0)
📝 libs/langgraph/tests/test_pregel.py (+12 -14)
📝 libs/langgraph/tests/test_pregel_async.py (+1 -3)

📄 Description

  • Use a simpler version of RunnableSequence without tracing serialization
  • Remove accepts_run_manager check in RunnableCallable
  • Remove creation of ChannelWrite dynamically every time conditional edge runs

🔄 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/1605 **Author:** [@nfcampos](https://github.com/nfcampos) **Created:** 9/4/2024 **Status:** ✅ Merged **Merged:** 9/4/2024 **Merged by:** [@nfcampos](https://github.com/nfcampos) **Base:** `nc/3sep/perf-optimizations` ← **Head:** `nc/4sep/perf-optimizations-ii` --- ### 📝 Commits (1) - [`de21b4f`](https://github.com/langchain-ai/langgraph/commit/de21b4f4c109e006ebaff3370ca1e0d948f0694f) Reduce from 9s to 4s on benchmark graph ### 📊 Changes **12 files changed** (+446 additions, -148 deletions) <details> <summary>View changed files</summary> 📝 `libs/langgraph/langgraph/graph/graph.py` (+14 -8) 📝 `libs/langgraph/langgraph/graph/state.py` (+9 -16) 📝 `libs/langgraph/langgraph/pregel/__init__.py` (+7 -15) 📝 `libs/langgraph/langgraph/pregel/algo.py` (+2 -2) 📝 `libs/langgraph/langgraph/pregel/read.py` (+10 -8) 📝 `libs/langgraph/langgraph/pregel/write.py` (+38 -56) 📝 `libs/langgraph/langgraph/utils/config.py` (+3 -1) 📝 `libs/langgraph/langgraph/utils/runnable.py` (+301 -20) 📝 `libs/langgraph/poetry.lock` (+10 -5) 📝 `libs/langgraph/tests/__snapshots__/test_pregel.ambr` (+39 -0) 📝 `libs/langgraph/tests/test_pregel.py` (+12 -14) 📝 `libs/langgraph/tests/test_pregel_async.py` (+1 -3) </details> ### 📄 Description - Use a simpler version of RunnableSequence without tracing serialization - Remove accepts_run_manager check in RunnableCallable - Remove creation of ChannelWrite dynamically every time conditional edge runs --- <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 17:46:27 -05:00
yindo closed this issue 2026-02-20 17:46:27 -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#2174