What is MemOS built with?

MemOS is a TypeScript repo built with FastAPI, uvicorn, anyio, starlette, fastmcp, and 8 more. It stores its data in Postgres and SQLite, reaches it with SQLAlchemy, signs people in with authlib, cryptography, itsdangerous, keyring, pyjwt, and 1 more, uses AI through langsmith, mcp, nltk, NVIDIA, nvidia-cudnn-cu12, and 17 more, makes its content with beautifulsoup4, defusedxml, docutils, et-xmlfile, jinja2, and 17 more, validates its data with annotated-types, beartype, email-validator, jsonschema, jsonschema-path, and 8 more, is tested with Pytest, Vitest, Testing Library, @playwright/test, and playwright, deploys with coloredlogs, concurrent-log-handler, fastapi-cloud-cli, volcengine-python-sdk, prometheus-client, and 1 more, and uses diskcache, psycopg2-binary, py-key-value-aio, pymysql, Redis, and 8 more. It has 447,484 written lines, with 480 test files.

Read on November 11, 2026, at commit 3a16e5f. Read it as markdown

The card says what it's built with. Its whole map opens it up: every part of the code and how it connects to the rest.

Frontend

Backend

Support

The base 5

Content 22

Validation 13

Testing 5

  • Vitest276 files
  • Pytest52 files
  • Testing Library11 files
  • @playwright/test5 files
  • playwright3 files

Deploy 6

Tools 14

  • click
  • cyclopts
  • fastapi-cli
  • pygments
  • Setuptools
  • Typer
  • typing-extensions13 files
  • docstring-parser
  • iniconfig
  • packaging
  • pluggy
  • pycparser
2 more
  • shellingham
  • watchfiles

Other 81

  • uuid16 files
  • @preact/signals7 files
  • NumPy7 files
  • pyyaml4 files
  • SemVer2 files
  • tqdm2 files
  • .ENV1 file
  • @deepseek-ai/dsh-timeout1 file
  • @hono/node-ws1 file
  • python-dateutil1 file
  • zustand1 file
  • attrs
69 more
  • certifi
  • dnspython
  • filelock
  • fsspec
  • hf-xet
  • joblib
  • networkx
  • nvidia-nccl-cu12
  • pandas
  • platformdirs
  • portalocker
  • pydantic-settings
  • requests-toolbelt
  • SciPy
  • SymPy
  • ujson
  • urllib3
  • websockets
  • requests13 files
  • python-dotenv9 files
  • jieba4 files
  • httpx3 files
  • cachetools2 files
  • tenacity2 files
  • node-pty1 file
  • schedule1 file
  • async-timeout
  • cffi
  • charset-normalizer
  • cobble
  • colorama
  • distro
  • exceptiongroup
  • flatbuffers
  • greenlet
  • grpcio
  • h11
  • h2
  • hpack
  • httpcore
  • httptools
  • httpx-sse
  • humanfriendly
  • hyperframe
  • idna
  • jaraco.context
  • jaraco.functools
  • jiter
  • jsonpatch
  • jsonpointer
  • more-itertools
  • mpmath
  • orjson
  • pathable
  • protobuf
  • pyperclip
  • python-multipart
  • pytz
  • regex
  • rignore
  • rpds-py
  • six
  • sniffio
  • sse-starlette
  • threadpoolctl
  • typing-inspection
  • tzdata
  • uvloop
  • zstandard

Other repos with a similar stack

Ordered by how much of its stack they share with MemOS, then by stars.

Frequently asked questions

What language is MemOS written in?
MemOS is written in TypeScript, with 447,484 written lines.
What framework does MemOS use?
MemOS is built with FastAPI, uvicorn, anyio, starlette, fastmcp, Preact, @deepseek-ai/cordis, Vite, React, react-router-dom, Electron, Express, and Hono.
What database does MemOS use?
MemOS stores its data in Postgres and SQLite and reaches it with SQLAlchemy.

The proof: src/memos/graph_dbs/postgres.py, line 1

What does MemOS sign people in with?
MemOS signs people in with authlib, cryptography, itsdangerous, keyring, pyjwt, and @aws-sdk/credential-providers.
Does MemOS use AI?
Yes. MemOS uses AI through langsmith, mcp, nltk, NVIDIA, nvidia-cudnn-cu12, onnxruntime, safetensors, tavily-python, tokenizers, triton, scikit-learn, torch, transformers, LangChain, Ollama, chonkie, Hugging Face, rank-bm25, sentence-transformers, OpenAI, deepseek-ai, and @aws-sdk/client-bedrock.
What services does MemOS use?
MemOS uses diskcache, psycopg2-binary, py-key-value-aio, pymysql, Redis, Neo4j, pika, pymilvus, Alibaba Cloud, Qdrant, @deepseek-ai/dsh-session, electron-store, and MCP.
Does MemOS have tests?
Yes. MemOS has 480 test files and is tested with Pytest, Vitest, Testing Library, @playwright/test, and playwright.
Which repos have a stack like MemOS's?

See inside your code, today.

Paste your repo and see it drawn, part by part: a medium one in under a minute.

Free, no account.

Or look inside SeeRepo's own →

Is it your repo? Add its badge, keep it up to date or take it down →