智能记忆层

Install

pip install widemem-ai[faiss]

The [faiss] extra installs the default local vector store. Plain pip install widemem-ai installs the core only; you'll need at least one vector backend ([faiss] or [qdrant]) before WideMemory() will work. Python 3.10+ required.

Install

pip install widemem-ai[mcp,sentence-transformers]

Quick Start

Five lines to a working memory system. Six if you count the import.

```python from widemem import WideMemory, MemoryConfig

memory = WideMemory()

Optional providers

pip install widemem-ai[anthropic]             # Claude LLM provider
pip install widemem-ai[ollama]                # Local LLM via Ollama
pip install widemem-ai[sentence-transformers] # Local embeddings (no API key needed)
pip install widemem-ai[qdrant]                # Qdrant vector store
pip install widemem-ai[mcp]                   # Model Context Protocol server
pip install widemem-ai[all]                   # Everything. You want it all? You got it.

---

Configuration

Most defaults are sane, so a minimal config is usually enough:

from widemem import WideMemory, MemoryConfig
from widemem.core.types import LLMConfig, ScoringConfig, YMYLConfig

config = MemoryConfig(
    llm=LLMConfig(provider="openai", model="gpt-4o-mini"),
    scoring=ScoringConfig(decay_rate=0.01),
    ymyl=YMYLConfig(enabled=True),
    history_db_path="~/.widemem/history.db",
)
memory = WideMemory(config)

Full reference for every field, default, and tradeoff: docs/configuration.md.

---

Set at config level (default for all queries)

mem = WideMemory(config=MemoryConfig(retrieval_mode="balanced"))

API Reference

Full method signatures, parameters, and return types: docs/api.md.

The most-used surface area:

---

Framework integrations

• #22 LangChain BaseChatMessageHistory adapter — drop-in conversation-history backend for LangChain chains and agents
• #23 LangChain BaseRetriever adapter — RAG-style retrieval from widemem in any LangChain chain
• #24 LangGraph BaseStore adapter — memory backend for stateful LangGraph agents