How do I fix ModuleNotFoundError?
Check which Python environment you're using. Activate your virtualenv and run pip list to verify the package is installed.
AI-native database for everything you remember.
JSON tables, semantic search, agent memory.
Zero core dependencies. Human-readable by design.
❯ neudb table create users
❯ neudb row insert users --data '{"username":"alice"}'
❯ neudb row list users
{'username': 'alice', 'id': '7e6b0543-...'}
❯
Search similar messages
Inject memory context
Store user + assistant turn
◆ Thought for 0.3s
42context = db.search_similar(query_vec)
43answer = llm.chat(messages + context)
43answer = llm.chat(messages + context)
Python import errors
sim=0.509 · [assistant] Check which Python environment you are using.
sim=0.412 · [user] How do I install packages in a venv?
POST /search
X-API-Key: ••••••••
{
"query": "Python import error",
"top_k": 5
}
→ 200 OK · 3 results
X-API-Key: ••••••••
{
"query": "Python import error",
"top_k": 5
}
→ 200 OK · 3 results
For developers
One engine. Every interface.
CLI, Python library, HTTP API, and agent loop — all through a single storage engine. Start building in seconds.
0
Core dependencies
<50ms
Indexed search
5
AI memory tables
from neudb.ai_schema import * db = init_ai_database("my_memory") alice = add_user(db, "alice", "alice@example.com") session = create_session(db, alice, "Debugging Python") add_message_with_embedding( db, session, "user", "How do I fix ModuleNotFoundError?" ) vec = embed_text("Python import errors") results = db.table("messages").search_similar( "embedding", vec, top_k=5 )
# Create and query tables neudb table create users neudb row insert users --data '{"username":"alice"}' neudb row list users # Semantic search neudb row search messages \ --field embedding \ --vector '[0.9, 0.1, 0.2]' # Agent with long-term memory neudb-agent --provider ollama --model llama3.2
# Create a user curl -X POST http://127.0.0.1:8000/users \ -H "Content-Type: application/json" \ -H "X-API-Key: $NEUDB_API_KEY" \ -d '{"username":"alice","email":"alice@example.com"}' # Semantic search curl -X POST http://127.0.0.1:8000/search \ -H "X-API-Key: $NEUDB_API_KEY" \ -d '{"query":"Python import error","top_k":5}'
24
tests passing in CI
0
core dependencies required
5
memory tables built in
Mission to make AI memory debuggable.
Latest news
All posts →Choose how to get started
Build on your own
Launch your agent or prototype with:
- Zero core dependencies
- CLI, library, and optional HTTP API
- Human-readable JSON you can cat
- Semantic search with text fallback
Run the full stack
API server and LLM agent with memory:
- FastAPI with API key auth
- Ollama or OpenAI agent loop
- Automatic message embedding
- Prompt injection guards built in