letta
25,510Stateful agents framework with memory, reasoning, and context management (formerly MemGPT).
Activity band details
Activity band A to F: a bin of a score that adds GitHub stars (log scale), a push to the upstream repository in the 12 months before `as_of`, the number of declared integration surfaces and, where the catalogue records them, citing papers and benchmark evaluations. It measures popularity and activity. It is not a level of evidence: no band means a paper cites the tool or that it was measured. Recency is judged as of 2026-11-22.
Computed band: C
(raw score 4.76).
Shown band: C
.
Facets:
documented_in=0evaluated_on=0integration_surfaces=1paper_citations=0recent_push=2stars=1.76
Source of truth: pipeline/maturity/score.py
Description
Letta (formerly MemGPT) introduces 'MemGPT-style' long-term memory for LLM agents: tiered context buffers, summarization, and explicit memory tools that let agents write to and retrieve from long-term storage.
Why it's included
Memory-management framework for stateful agents. Tag basis (2026-11-20 tag audit, research/fm_tag_audit/final_table.csv): scope_creep was kept for permission modes and human approval of tool calls; context_pollution was removed because memory blocks and compaction are documented for persistence and context limits, not as a mitigation of polluted context.
Addresses failure modes
These tags are the catalog's claim about what letta is built to catch or prevent. A tag is not a test result. SAICA-Bench has not measured letta, so none of these tags is tested here.
- Scope creep — Actions outside declared task boundary.