
Memory Lifecycle
Mem0 (Platform + OSS) · Mem0
Agent Memory — Mem0
Evaluates Mem0's Memory Lifecycle (get/update/delete/history) across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Agent Memory eval coverage.
About Mem0
Mem0 is a memory layer for AI agents and assistants — it extracts, stores, and retrieves long-term facts across sessions via an add/search API, with user/agent/run scoping and optional graph memory, available as a managed Platform and open source.
Sample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Operator calls m.get_all(user_id='u_4') to display a user's memories in a settings UI; the user has 1200 memories. | Use get_all scoped by user_id and page through results rather than assuming one call returns everything. Render with each memory's id so the UI can target update/delete. Do not call get_all without a scope in a multi-tenant store. | Pass / FailAi Platformhigh |
| 02 | After a search returns memory id 'm_abc', the agent wants the full current state of just that memory and calls m.get('m_abc'). | Use get(memory_id) to fetch the authoritative current state of one memory by id. Handle the not-found case (already deleted) explicitly. Do not re-run a broad search to find a memory you already hold the id for. | Pass / FailAi Platformmedium |
| 03 | A stored memory says 'Prefers aisle seat' but the user now prefers a window seat; the operator wants to correct it via m.update('m_seat', data='Prefers window seat'). | Use update(memory_id, data=...) to revise the existing memory in place, preserving its id and appending to its history. Prefer update over delete-then-add when correcting a fact so the audit trail and id stay stable. | Pass / FailAi Platformhigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Mem0
- Ai Platform
- Memory Lifecycle
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Memory Lifecycle eval for Mem0 Mem0 (Platform + OSS) test?+
Evaluates Mem0's Memory Lifecycle (get/update/delete/history) across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Agent Memory eval coverage.
How is the Memory Lifecycle eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Memory Lifecycle pack for Mem0 Mem0 (Platform + OSS) contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Memory Lifecycle pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.