Agent Frameworks
Building AI agents with LangChain, LangGraph, CrewAI, LlamaIndex, the OpenAI Agents SDK, Google ADK, Mastra and more, including adding long-term memory to each.
Every agent framework ships some notion of memory, and almost none of them keep it past the end of a session. LangGraph checkpoints a thread and the OpenAI Agents SDK hands you a session object; what happens when the same user returns next week is left to you.
These articles work through one framework at a time: how its state and memory model behaves, where it stops, and how to add persistent memory without rewriting your agent. The Maximem Synap integration guides here cover LangChain, LangGraph, LlamaIndex, CrewAI, AutoGen, Semantic Kernel, Mastra, the Vercel AI SDK, Google ADK, Pydantic AI, the Claude Agent SDK, the OpenAI Agents SDK, LiveKit and Pipecat.
Start here: Synap integrations · How agent memory works
Harness And Agents
Every LLM call is stateless, and a model on its own cannot act; the software that turns it into an agent is the harness.
How to Add Conversational Memory to a LangChain App (2026 Guide)
To add conversational memory to a LangChain app today, pass a checkpointer to create_agent and reuse one thread_id per conversation; to remember a person across conversations, add a LangGraph store or plug in Maximem Synap through the maximem-synap-langchain package.
How to Add Memory to a Pipecat Voice Agent
Pipecat's LLMContext and auto summarization manage one pipeline run; for recall across calls its docs list Mem0MemoryService. Synap ships the same pattern as two frame processors, SynapMemoryProcessor and SynapRecorder, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
LiveKit Agents Memory: What Is the Best Memory System for Voice AI?
LiveKit Agents keeps ChatContext and userdata for one call and has no long-term memory store of its own. A voice agent needs caller history loaded before the first reply. Synap preloads it and records each turn, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
Semantic Kernel Memory: How to Add Memory to an Agent
Semantic Kernel is now Microsoft Agent Framework, and many agents still run on it. ChatHistory and its reducers manage one conversation; per-user memory across sessions is yours to add. Synap's Python plugin adds it, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
AutoGen Memory: How to Add Memory to AutoGen Agents
AutoGen is in maintenance mode, and its AgentChat Memory protocol stores what your code adds. Synap adds user-scoped search and store tools, at 92% LongMemEval, 93.2% on LoCoMo accuracy, plus a separate package for Microsoft Agent Framework.
Maximem Synap's Agent Memory Now Available for CrewAI
Maximem Synap's persistent agent memory now integrates with CrewAI. Per-user persistence, entity resolution, and 92% LongMemEval, 93.2% on LoCoMo accuracy for multi-agent production workflows.
Google ADK Memory: How to Add Persistent Memory to an Agent
Google ADK gives you Session, State and a MemoryService; its persistent memory options run on Google Cloud. Synap adds per-user memory as two tools that work wherever the agent runs, with entity resolution and 92% LongMemEval, 93.2% on LoCoMo accuracy.
Maximem Synap's Agent Memory Now Available for Mastra
Maximem Synap's persistent agent memory now integrates with Mastra. Per-user persistence, entity resolution, and 92% LongMemEval, 93.2% on LoCoMo accuracy for TypeScript-native production agents.
Maximem Synap's Agent Memory Now Available for Vercel AI SDK
Maximem Synap's persistent agent memory now integrates with Vercel AI SDK. Per-user persistence, entity resolution, and 92% LongMemEval, 93.2% on LoCoMo accuracy for JavaScript/TypeScript production agents.
Claude Agent SDK Memory: What Memory Layer Works With It?
The Claude Agent SDK saves each session's transcript and loads CLAUDE.md files, but both are keyed to a session or a directory and not to your end user. Synap adds per-user memory and entity resolution, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
Pydantic AI Memory: How to Add Memory to an Agent
Pydantic AI core has no memory store: an agent remembers only the messages you pass back through message_history, and storing them is your job. Synap adds per-user memory through typed dependencies and two tools, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
LlamaIndex Memory: How to Add Persistent Memory to an Agent
LlamaIndex's Memory class keeps chat history per session_id plus optional long-term blocks, in an in-memory SQLite database unless you pass a database URI. A per-user scope is left to you. Synap adds user-scoped memory, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
OpenAI Agents SDK Memory: Which Memory System Should You Use?
OpenAI Agents SDK sessions store and replay one conversation's history per session ID. They do not carry facts about a user into a new session. Synap adds per-user memory and entity resolution as two tools, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
How to Add Persistent Memory to a LangGraph Agent
LangGraph checkpointers save state per thread and its Store holds data across threads, but your code decides what to write and how to update it. Synap's SynapStore adds user-scoped memory with entity resolution, at 92% LongMemEval, 93.2% on LoCoMo accuracy.
LangChain Long-Term Memory: Remembering Users Across Sessions
LangChain keeps one conversation with a checkpointer, scoped to a single thread_id, and keeps data across conversations in a LangGraph store under a namespace you design. What to save, how a changed fact replaces an old one and how users stay separate is left to your code.