Skip to main content
StremAI
Sign InSign Up
Migration Guide

Migrate from Mem0 to StremAI

Switch in under 10 minutes. This guide maps every Mem0 API call to its StremAI equivalent.

Step 1: Install

Replace the Mem0 package:

# Remove Mem0
pip uninstall mem0ai

# Install StremAI
pip install agentbay
Step 2: Initialize

Mem0 (before)

from mem0 import Memory

m = Memory()
# or with API key:
m = Memory.from_config({
    "api_key": "m0-xxx"
})

StremAI (after)

from agentbay import AgentBay

ab = AgentBay()  # local mode
# or with API key:
ab = AgentBay(
    api_key="ab_live_xxx",
    project_id="my-project"
)

StremAI with no arguments starts in local mode (SQLite). No signup needed to get started.

Step 3: Store Memories

Mem0

m.add(
    "The user prefers dark mode",
    user_id="user1",
    metadata={"category": "preference"}
)

StremAI

# Drop-in (Mem0-compatible)
ab.add(
    "The user prefers dark mode",
    user_id="user1"
)

# Or structured, for better search
ab.store(
    "The user prefers dark mode",
    title="User prefers dark mode",
    type="PATTERN",
    tags=["preference", "ui"]
)

ab.add() is a drop-in replacement for m.add() — it auto-detects the type and extracts a title. ab.store() lets you set title, type (PATTERN, PITFALL, DECISION, etc.), and tags explicitly. Both return a dict: {'id': '...', 'deduplicated': False}.

Step 4: Recall Memories

Mem0

results = m.search(
    "dark mode",
    user_id="user1"
)
for r in results:
    print(r["memory"])

StremAI

results = ab.recall("dark mode")
# or drop-in: ab.search("dark mode", user_id="user1")

for entry in results:
    print(f"{entry['title']}: {entry['content']}")
    print(f"  confidence: {entry['confidence']}")

ab.recall() returns a plain list of dicts with confidence scores (computed at query time with decay). ab.search() is a Mem0-compatible alias with the same user_id scoping.

Step 5: Auto-Memory Chat

Mem0

# Mem0 doesn't have a built-in
# chat wrapper. You must manually:
# 1. Search for relevant memories
# 2. Inject into system prompt
# 3. Call your LLM
# 4. Store the response

StremAI

reply = ab.chat(
    [{"role": "user", "content": "How does auth work?"}],
    provider="openai",   # or "auto" / "anthropic" / "ollama" ...
    model="gpt-4o",
)
# Auto: recall → inject → LLM → store
# Returns the raw provider response object
print(reply.choices[0].message.content)
Full API Mapping
Mem0StremAINotes
m.add(text, user_id)ab.add(text, user_id)Drop-in; or ab.store(content, title=, type=)
m.search(query, user_id)ab.search(query, user_id)Drop-in alias for ab.recall(query)
m.get_all(user_id)ab.health() / LocalMemory().export()Health stats, or a full local dump
m.delete(memory_id)ab.forget(id)Soft delete (archived, recoverable)
m.update(memory_id, data)ab.store(...)Auto-dedup updates the existing entry
m.history(memory_id)Brain time machine (MCP / cloud)Snapshots + rollback
(no equivalent)ab.verify(id)Reset confidence decay
(no equivalent)memory_compact (MCP tool)Archive stale, merge dupes
(no equivalent)ab.chat(messages)Auto-memory LLM wrapper
What You Gain by Switching
  • Better search: 4-strategy RRF fusion instead of vector-only
  • Confidence decay: Old memories fade, frequently-used ones stay strong
  • Poison detection: 20+ patterns block prompt injection in stored memories
  • Local mode: Work offline, sync to cloud when ready
  • brain.chat(): Auto-memory wrapping for any LLM in one line
  • Memory tiers: Working (24h) → episodic → semantic → procedural (365d)
  • Multi-agent teams: Shared knowledge with role-based permissions
  • MCP tools: Works with Claude Code, Codex, Cursor, any MCP client (tool count varies by transport)

Questions about migrating?

Start with local mode to test without commitment.

QuickstartFull Comparison