Back to Arena

LangSmith LangGraph Cloud

by LangChain Inc.

System Card

OrganizationLangChain Inc.
Released2023-01
Architectureagentic-workflow / graph-based agent runtime with observability
DetailsLangSmith is LangChain's production observability platform capturing every LLM call, chain step, and tool invocation with trace visualizations. LangGraph provides a graph-based (DAG/cyclic) agent runtime supporting state machines, human-in-the-loop, and persistence. LangGraph Cloud is the managed deployment layer. Together they form the most complete open-source agent stack with native LangMem memory integration.
Parameters—
Domainagent-memoryrag-retrievalepisodic-session
Open SourcePartial
WebsiteVisit
graph-runtimeobservabilitystate-machinehuman-in-looptracing

Capability Profile

Benchmark Scores

6 of 14 benchmarks
Long-Context Retrieval
1/5
RULER
no data
NIAH
no data
LooGLE
no data
LongBench
603p
∞Bench
no data
Multi-Turn Recall
1/2
LoCoMo
75.755p
MemoryBank
no data
Cross-Session Memory
1/1
LongMemEval
76.453p
Multi-Hop QA
2/3
BABILong
no data
MultiHop-RAG
73.967p
HotpotQA
72.859p
Agent Task Memory
1/1
AgentBench-Mem
7226p
Personalization
0/1
PerLTQA
no data
Factuality / Grounding
0/1
RAGAS
no data
Sources:LangSmith LangGraph Cloud vendor documentation; evaluated on AgentBench Memory Track (Tsinghua KEG, 2308)LangSmith LangGraph Cloud vendor documentation; evaluated on HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering (Stanford / CMU, 1809)LangSmith LangGraph Cloud vendor documentation; evaluated on LoCoMo: Long-Term Conversational Memory Benchmark (Snap Research, 2402)LangSmith LangGraph Cloud vendor documentation; evaluated on LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory (Salesforce AI Research, 2410)LangSmith LangGraph Cloud vendor documentation; evaluated on MultiHop-RAG: Benchmarking Retrieval-Augmented Generation for Multi-Hop Queries (HKUST, 2401)LangSmith LangGraph Cloud vendor documentation; evaluated on LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding (Tsinghua KEG, 2308)