Content
<!--
This README is the SINGLE SOURCE OF TRUTH, maintained in the private monorepo at
public_sdk/README.md and synced to the public repo root by scripts/sync_to_public.sh.
DO NOT edit the copy in maximem-ai/maximem_synap_sdk directly: changes there are
overwritten on the next sync. Edit here instead.
The integration table below is generated by scripts/gen_readme_integrations.py.
-->
<p align="center">
<picture>
<source media="(prefers-color-scheme: dark)" srcset="assets/banner-dark.png">
<source media="(prefers-color-scheme: light)" srcset="assets/banner-light.png">
<img src="assets/banner-light.png" alt="Synap: memory for production AI agents" width="100%" />
</picture>
</p>
<h3 align="center">Your agents forget between conversations. Synap makes them remember.</h3>
<p align="center">
<a href="https://docs.maximem.ai"><strong>Docs</strong></a> ·
<a href="#60-second-quickstart"><strong>Quickstart</strong></a> ·
<a href="https://github.com/maximem-ai/memory_and_context_eval_harness"><strong>Evals (reproduce our numbers)</strong></a> ·
<a href="https://github.com/maximem-ai/maximem_synap_sdk/discussions"><strong>Discussions</strong></a> ·
<a href="https://synap.maximem.ai"><strong>Dashboard</strong></a>
</p>
<p align="center">
<a href="https://pypi.org/project/maximem-synap"><img src="https://img.shields.io/pypi/v/maximem-synap?style=flat-square&color=blue&label=pypi" alt="PyPI" /></a>
<a href="https://pypi.org/project/maximem-synap"><img src="https://img.shields.io/pypi/dm/maximem-synap?style=flat-square&color=blue" alt="PyPI Downloads" /></a>
<a href="https://www.npmjs.com/package/@maximem/synap-js-sdk"><img src="https://img.shields.io/npm/v/@maximem/synap-js-sdk?style=flat-square&color=blue&label=npm" alt="npm" /></a>
<a href="https://pypi.org/project/maximem-synap"><img src="https://img.shields.io/pypi/pyversions/maximem-synap?style=flat-square" alt="Python versions" /></a>
<a href="LICENSE"><img src="https://img.shields.io/badge/license-Apache%202.0-blue?style=flat-square" alt="License" /></a>
<a href="https://github.com/maximem-ai/memory_and_context_eval_harness"><img src="https://img.shields.io/badge/evals-open%20harness-2ea44f?style=flat-square" alt="Open eval harness" /></a>
</p>
---
## 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](https://github.com/maximem-ai/memory_and_context_eval_harness)), fetches context in ~15ms P50 (internal measurement) before your agent asks, and ships native packages for 24 agent frameworks.
<p align="center">
<strong>LangChain · LangGraph · LlamaIndex · CrewAI · AutoGen · Haystack · Google ADK · OpenAI Agents · Semantic Kernel · Pydantic AI · Agno · LiveKit · Pipecat · Claude Agent · Mastra · Vercel AI SDK · NeMo Agent Toolkit · Microsoft Agent Framework</strong>
</p>
### What's open, what's hosted
| | This repo (Apache 2.0) | Synap Cloud (managed) |
|---|---|---|
| Python + JS SDKs, 24 framework packages, MCP adapter | ✅ open source | uses them |
| Memory engine: ingestion, entity resolution, retrieval, anticipation | — | ✅ runs here |
| What you need | `pip install maximem-synap` | an [API key](https://www.maximem.ai/synap). Free Trial plan: 12,500 credits/month, 1 agent, no card required ([pricing](https://www.maximem.ai/synap/pricing)) |
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.
| Benchmark | Synap accuracy | Scope |
|---|---|---|
| **LongMemEval** | **92.0%** (460 / 500) | Full set, 6 categories |
| **LoCoMo** | **93.2%** | Cat 1–4, adversarial excluded |
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 →](https://github.com/maximem-ai/eval_benchmark_runs_output/blob/main/METHODOLOGY.md)
Category breakdown, methodology and every scope decision → [eval_benchmark_runs_output](https://github.com/maximem-ai/eval_benchmark_runs_output)
---
## Install
```bash
# 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.
<details>
<summary><strong>Claude Code</strong></summary>
```bash
git clone --depth 1 https://github.com/maximem-ai/maximem_synap_sdk /tmp/synap && cp -r /tmp/synap/skills/synap ~/.claude/skills/synap
```
</details>
<details>
<summary><strong>Codex</strong></summary>
```bash
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
```
</details>
<details>
<summary><strong>Cursor / any MCP client</strong></summary>
Add the Synap MCP server URL and your API key from the [dashboard](https://synap.maximem.ai). See [`packages/mcps/synap-mcp-server/`](packages/mcps/synap-mcp-server/).
</details>
---
## 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](https://www.maximem.ai/synap). There's no local server to run: memory processing happens in Synap's cloud, not in this package.
```python
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())
```
<details>
<summary><strong>JavaScript / TypeScript</strong></summary>
```javascript
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);
```
</details>
---
## 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
```bash
pip install maximem-synap maximem-synap-langchain
```
```python
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
```bash
pip install maximem-synap maximem-synap-crewai
```
```python
from synap_crewai import SynapStorageBackend
crew = Crew(
agents=[...], tasks=[...],
memory=True,
storage=SynapStorageBackend(sdk=sdk, user_id="alice"),
)
```
### LlamaIndex
```bash
pip install maximem-synap maximem-synap-llamaindex
```
```python
from synap_llamaindex import SynapChatMemory
memory = SynapChatMemory(sdk=sdk, user_id="alice")
agent = ReActAgent.from_tools(tools, memory=memory)
```
### All integrations
<!-- BEGIN integrations (generated by scripts/gen_readme_integrations.py) -->
| Framework | Package | Install |
|---|---|---|
| LangChain | [`maximem-synap-langchain`](packages/integrations/synap-langchain/) | `pip install maximem-synap-langchain` |
| LangGraph | [`maximem-synap-langgraph`](packages/integrations/synap-langgraph/) | `pip install maximem-synap-langgraph` |
| LlamaIndex | [`maximem-synap-llamaindex`](packages/integrations/synap-llamaindex/) | `pip install maximem-synap-llamaindex` |
| CrewAI | [`maximem-synap-crewai`](packages/integrations/synap-crewai/) | `pip install maximem-synap-crewai` |
| AutoGen | [`maximem-synap-autogen`](packages/integrations/synap-autogen/) | `pip install maximem-synap-autogen` |
| Haystack | [`maximem-synap-haystack`](packages/integrations/synap-haystack/) | `pip install maximem-synap-haystack` |
| Google ADK | [`maximem-synap-google-adk`](packages/integrations/synap-google-adk/) | `pip install maximem-synap-google-adk` |
| OpenAI Agents SDK | [`maximem-synap-openai-agents`](packages/integrations/synap-openai-agents/) | `pip install maximem-synap-openai-agents` |
| Semantic Kernel | [`maximem-synap-semantic-kernel`](packages/integrations/synap-semantic-kernel/) | `pip install maximem-synap-semantic-kernel` |
| Pydantic AI | [`maximem-synap-pydantic-ai`](packages/integrations/synap-pydantic-ai/) | `pip install maximem-synap-pydantic-ai` |
| Agno | [`maximem-synap-agno`](packages/integrations/synap-agno/) | `pip install maximem-synap-agno` |
| LiveKit Agents | [`maximem-synap-livekit-agents`](packages/integrations/synap-livekit-agents/) | `pip install maximem-synap-livekit-agents` |
| Pipecat | [`maximem-synap-pipecat`](packages/integrations/synap-pipecat/) | `pip install maximem-synap-pipecat` |
| Claude Agent (Python) | [`maximem-synap-claude-agent`](packages/integrations/synap-claude-agent/) | `pip install maximem-synap-claude-agent` |
| Claude Agent (TypeScript) | [`@maximem/synap-claude-agent`](packages/integrations/synap-claude-agent-ts/) | `npm i @maximem/synap-claude-agent` |
| Mastra | [`@maximem/synap-mastra`](packages/integrations/synap-mastra/) | `npm i @maximem/synap-mastra` |
| Vercel AI SDK | [`@maximem/synap-vercel-adk`](packages/integrations/synap-vercel-adk/) | `npm i @maximem/synap-vercel-adk` |
| Vercel eve | [`@maximem/synap-eve`](packages/integrations/synap-eve/) | `npm i @maximem/synap-eve` |
| NVIDIA NeMo Agent Toolkit | [`maximem-synap-nemo-agent-toolkit`](packages/integrations/synap-nemo-agent-toolkit/) | `pip install maximem-synap-nemo-agent-toolkit` |
| Microsoft Agent Framework | [`maximem-synap-microsoft-agent`](packages/integrations/synap-microsoft-agent/) | `pip install maximem-synap-microsoft-agent` |
| Strands Agents | [`maximem-synap-strands-agents`](packages/integrations/synap-strands-agents/) | `pip install maximem-synap-strands-agents` |
| CAMEL-AI | [`maximem-synap-camel-ai`](packages/integrations/synap-camel-ai/) | `pip install maximem-synap-camel-ai` |
| Smolagents | [`maximem-synap-smolagents`](packages/integrations/synap-smolagents/) | `pip install maximem-synap-smolagents` |
| Deepagents | [`maximem-synap-deepagents`](packages/integrations/synap-deepagents/) | `pip install maximem-synap-deepagents` |
| Remote MCP Server | [`maximem-synap-mcp-server`](packages/mcps/synap-mcp-server/) | `pip install maximem-synap-mcp-server` |
<!-- END 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](https://synap.maximem.ai)), or self-host the adapter from source.
→ Source & self-hosting: [`packages/mcps/synap-mcp-server/`](packages/mcps/synap-mcp-server/)
---
## Deep dives
Understand the system before building on it:
- 📘 **[Why we built Synap](https://www.maximem.ai/blog/why-we-built-synap)**: the problem with current AI memory systems
- ⚙️ **[How Synap works under the hood](https://www.maximem.ai/blog/how-maximem-synap-works)**: architecture, retrieval pipeline, and design decisions
- 📊 **[Benchmark results](https://www.maximem.ai/blog/synap-benchmark-results)**: 92.0% on LongMemEval, 93.2% on LoCoMo, methodology, and reproducibility
- 🧪 **[Evaluation harness](https://github.com/maximem-ai/memory_and_context_eval_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?](#using-a-coding-agent).
- **[Synap skill for Claude Code](skills/synap/)**: SDK setup, scoping (User/Customer/Client), ingestion, retrieval, and one-page wiring guides for all supported frameworks.
- **[Synap skill for Codex](skills/synap-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](https://www.maximem.ai/synap)
---
## Resources & community
- 📖 [Documentation](https://docs.maximem.ai)
- 🚀 [Dashboard](https://synap.maximem.ai)
- 💬 [GitHub Discussions](https://github.com/maximem-ai/maximem_synap_sdk/discussions): questions, ideas, show what you built
- 🔒 [Security policy](SECURITY.md)
- 🤖 [AGENTS.md](AGENTS.md) and [llms.txt](llms.txt) for coding agents
- 𝕏 [Twitter / X](https://x.com/maximem_ai)
- 💼 [LinkedIn](https://www.linkedin.com/company/maximem-ai)
---
## 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](https://github.com/maximem-ai/maximem_synap_sdk/issues), and questions go to [Discussions](https://github.com/maximem-ai/maximem_synap_sdk/discussions). See [CONTRIBUTING.md](CONTRIBUTING.md) for how to run the code locally and what a new integration needs.
---
## License
Apache 2.0 (see [LICENSE](LICENSE)). Each package declares its own license in its metadata; the core Python and JS SDKs and the Vercel AI SDK integration are MIT.
---
<p align="center">
Built by <a href="https://www.maximem.ai"><strong>Maximem AI</strong></a>
</p>
Connection Info
You Might Also Like
No recommendations available