Synap Memory

The memory layer for production AI agents

Synap gives your agents memory that lasts across conversations, sessions and devices. It scores 92.0% on LongMemEval and 93.2% on LoCoMo (gpt-5-mini answer and judge, reproduce with our open harness), fetches context in ~15ms P50 (internal measurement) before your agent asks, and ships native packages for 24 agent frameworks.

What's open, what's hosted

Every plan, including the free one, gets every Synap capability. Plans differ on volume and support, not features.

Benchmarks

Synap's scores on the two standard long-term memory benchmarks, produced with our open-source harness on the official dataset releases.

Other vendors publish their own numbers under different answer models and judges. Our harness ships adapters for Mem0, Zep and Supermemory, so you can run them side by side yourself. How scores differ across setups →

Category breakdown, methodology and every scope decision → eval_benchmark_runs_output

Install

# Python
pip install maximem-synap

# JavaScript / TypeScript
npm install @maximem/synap-js-sdk

Using a coding agent?

Paste this into Claude Code, Codex or Cursor:

Add Synap memory to this project. Run pip install maximem-synap (or npm i @maximem/synap-js-sdk), then install the Synap skill from https://github.com/maximem-ai/maximem_synap_sdk/tree/main/skills/synap and follow it to wire memory into the agent framework this repo uses. Ask me for my Synap API key when you need it.

git clone --depth 1 https://github.com/maximem-ai/maximem_synap_sdk /tmp/synap && cp -r /tmp/synap/skills/synap ~/.claude/skills/synap

git clone --depth 1 https://github.com/maximem-ai/maximem_synap_sdk /tmp/synap && mkdir -p ~/.agents/skills && cp -r /tmp/synap/skills/synap-codex ~/.agents/skills/synap

Add the Synap MCP server URL and your API key from the dashboard. See packages/mcps/synap-mcp-server/.

60-second quickstart

Your agent forgets. Synap remembers across conversations, sessions, and devices.

The SDK connects to the hosted Synap cloud service, so you'll need an API key. There's no local server to run: memory processing happens in Synap's cloud, not in this package.

import asyncio
from maximem_synap import MaximemSynapSDK

sdk = MaximemSynapSDK(api_key="your-api-key")

async def main():
    await sdk.initialize()

    # Monday's standup
    await sdk.conversation.record_message(
        conversation_id="mon-standup",
        user_id="alice",
        role="user",
        content="I'm migrating our auth service to OAuth2 this sprint.",
    )

    # Friday: completely different conversation, same user
    context = await sdk.fetch(
        conversation_id="fri-review",
        user_id="alice",
        search_query=["what is alice working on?"],
    )

    print(context.formatted_context)
    # → "Alice is migrating the auth service to OAuth2 this sprint."

asyncio.run(main())

const { createClient } = require('@maximem/synap-js-sdk');

const client = createClient({ apiKey: 'your-api-key' });
await client.init();

// Record
await client.conversation.recordMessage({
    conversationId: 'mon-standup',
    userId: 'alice',
    role: 'user',
    content: "I'm migrating our auth service to OAuth2 this sprint.",
});

// Fetch later, anywhere
const context = await client.fetchUserContext({
    userId: 'alice',
    query: 'what is alice working on?',
});

console.log(context.formattedContext);

How it works: anticipatory memory

Most memory layers wait for the agent to ask. Synap works out what the agent will need next and has it ready. Six parts make that work.

🎯 Anticipatory Retrieval

Synap pre-fetches context before your agent requests it. ~15ms P50 latency in production (Maximem internal measurement). For voice AI agents, this is the difference between natural conversation and awkward pauses.

🔗 Entity Resolution

When a user says "my manager" in turn 3 and "Sarah" in turn 12, Synap resolves them automatically. Cross-session, cross-conversation, without the agent doing any work.

⏳ Temporal Awareness

Context from 30 minutes ago and context from 30 days ago should not carry equal weight. Synap applies temporal decay and relevance scoring so your agent surfaces the right information at the right time.

🧠 Conscious Forgetting

When a user says "ignore what I said about the budget," Synap processes that as a retraction, not just more context to store. Contradiction handling is built into the pipeline.

🏗️ Custom Memory Architectures

No universal memory model. Synap builds customized memory architectures per use case. Customer support agents and voice AI agents need different context strategies. Synap handles both.

🏢 Multi-Tenant Scoping

Built for B2B from day one. Memory is scoped across a four-level hierarchy:

client          → shared knowledge across your entire platform
  └── customer  → per-company context (multi-tenant B2B)
        └── user       → per-user memory and preferences
              └── conversation → in-session history

One fetch() call merges all relevant scopes in parallel.

Framework integrations

Installable packages, not code snippets. Deep framework surfaces with callbacks, graph nodes, retrievers, memories, and plugins.

LangChain

pip install maximem-synap maximem-synap-langchain

from maximem_synap import MaximemSynapSDK
from synap_langchain import SynapChatMessageHistory
from langchain_openai import ChatOpenAI
from langchain_core.runnables.history import RunnableWithMessageHistory

sdk = MaximemSynapSDK(api_key="your-api-key")
await sdk.initialize()

chain = RunnableWithMessageHistory(
    ChatOpenAI(),
    lambda session_id: SynapChatMessageHistory(
        sdk=sdk, conversation_id=session_id, user_id="alice",
    ),
)

CrewAI

pip install maximem-synap maximem-synap-crewai

from synap_crewai import SynapStorageBackend

crew = Crew(
    agents=[...], tasks=[...],
    memory=True,
    storage=SynapStorageBackend(sdk=sdk, user_id="alice"),
)

LlamaIndex

pip install maximem-synap maximem-synap-llamaindex

from synap_llamaindex import SynapChatMemory

memory = SynapChatMemory(sdk=sdk, user_id="alice")
agent = ReActAgent.from_tools(tools, memory=memory)

All integrations

MCP server

Give no-code and low-code agents (Gumloop, n8n, and any MCP-compatible client) persistent memory with nothing but an MCP URL and your Synap API key. No SDK, no code. The server exposes Synap's memory as four MCP tools (log_exchange, check_memory_status, recall_context, list_recent_memories) over Streamable HTTP.

It's a stateless adapter over the hosted Synap API: each call maps to one REST operation and your Bearer token is forwarded verbatim, so there's no separate backend to run. Use the managed endpoint (grab the URL and token from your dashboard), or self-host the adapter from source.

→ Source & self-hosting: packages/mcps/synap-mcp-server/

Deep dives

Understand the system before building on it:

• 📘 Why we built Synap: the problem with current AI memory systems
• ⚙️ How Synap works under the hood: architecture, retrieval pipeline, and design decisions
• 📊 Benchmark results: 92.0% on LongMemEval, 93.2% on LoCoMo, methodology, and reproducibility
• 🧪 Evaluation harness: run LoCoMo and LongMemEval yourself against Synap, Mem0, Zep and Supermemory

Agent skills

Drop-in instructions that teach coding agents how to wire Synap into your codebase. Install steps are in Using a coding agent?.

• Synap skill for Claude Code: SDK setup, scoping (User/Customer/Client), ingestion, retrieval, and one-page wiring guides for all supported frameworks.
• Synap skill for Codex: the same guide, packaged for Codex (~/.agents/skills/synap).

Requirements

• Python SDK: Python 3.11+
• JavaScript SDK: Node 20+ (Python 3.11+ for the bridge layer)
• A Synap API key: get one at maximem.ai

Resources & community

• 📖 Documentation
• 🚀 Dashboard
• 💬 GitHub Discussions: questions, ideas, show what you built
• 🔒 Security policy
• 🤖 AGENTS.md and llms.txt for coding agents
• 𝕏 Twitter / X
• 💼 LinkedIn

Contributing

This repo is a published mirror of Maximem's monorepo, so everything under packages/ is overwritten on each sync and pull requests against it are closed. Bug reports, gaps, and new-framework requests are very welcome as issues, and questions go to Discussions. See CONTRIBUTING.md for how to run the code locally and what a new integration needs.

License

Apache 2.0 (see LICENSE). Each package declares its own license in its metadata; the core Python and JS SDKs and the Vercel AI SDK integration are MIT.