A side-by-side of two agent memory for building AI agents — live GitHub data, languages, and what each is best at.
Short answer: Letta (MemGPT) leads Zep vs Letta (MemGPT) by community traction (★ 24k vs ★ 4.8k). Pick Zep for temporal memory; pick Letta (MemGPT) for persistent agents.
✓ Live data verified
| Zep | Letta (MemGPT) | |
|---|---|---|
| GitHub stars | ★ 4.8k | ★ 24k |
| Language | Go | Python |
| Category | Agent memory | Agent memory |
| Best for | temporal memory | persistent agents |
| Repository | getzep/zep | letta-ai/letta |
Zep and Letta (MemGPT) are both credible choices. By community traction, Letta (MemGPT) leads (★ 24k). Pick Zep for temporal memory; pick Letta (MemGPT) for persistent agents.
Both are credible agent memory. By community traction Letta (MemGPT) leads (★ 24k). Pick Zep for temporal memory; pick Letta (MemGPT) for persistent agents.
Zep is Long-term memory store for agents with a temporal knowledge graph of facts and their validity over time.. Letta (MemGPT) is Stateful agents with long-term memory and self-editing context, evolved from the MemGPT research..
Letta (MemGPT) has more — ★ 24k vs ★ 4.8k (live counts).
Often yes — many teams combine agent memory. Check each tool's docs for interop; they solve overlapping but not identical problems.
Zep is primarily Go; Letta (MemGPT) is primarily Python.
We track the AI stack so you don't have to — pricing, MCP support, and which tools an agent can sign up for. Free.