Vocabulary drift The words the memory field uses to describe itself are not static:
"RAG" arrived in 2020, "agentic memory" became a phrase in 2024,
"knowledge-graph" peaked then receded. Tracking which terms enter,
spread, or fade — at the concept level rather than the system level —
catches recipe shifts the lineage view (which follows specific
systems) misses. A new vocabulary cluster is usually the first
signal that practitioners agree a new pattern matters.
27 curated memory terms tracked across
45 quarters · 912 of 912
records in scope after filters. ← back to analyses hub
How this view counts One record, one mention. If a term appears 12 times
in the same record's desc + claims we count
it once — we want "did this artefact adopt the vocabulary?", not
"how marketing-heavy was the copy?".Word-boundary, case-insensitive. "graph" matches
"graph" and "graphs" but not "graphical"; "RAG" matches "RAG-fusion".Bucketed by created date. Same parser as
the timeline view; records with year-only dates land in Q1 of that
year. Records with unparseable dates are excluded from the chart but
still contribute to co-occurrence and first-appearance counts.Curated 27-term list. Five thematic groups (cognitive
/ mechanism / agent-era / architectural / emerging). Extend via the
helper module + a DECISIONS.md note. Anything outside this list shows
up only as novel-vocab share in the ratio chart below — we
don't claim to track every term, just the ones we curate.Small-N badge. A growth % is flagged when fewer than
5 records sit on either side of the YoY ratio —
without this, "2 → 13" reads as "+550%" and dominates the eye. See
the "so what" callout for how to read flagged numbers.How to read it Fast-rising terms = emerging recipes. A term whose
cumulative arrival curve elbowed up in the last 4–6 quarters has
crossed from novelty into common vocabulary; the systems mentioning
it are usually converging on a new pattern. Open the drill panel
to see which.
Falling terms — two reads. Either the term is being
absorbed (e.g. "knowledge-graph" being subsumed by "graph-RAG"
where the same shape lives under a new label) or genuinely fading.
Compare against the cumulative chart: if cumulative is still rising,
the term is alive but quieter; if cumulative has plateaued for ≥4
quarters, it's actually receding.
Small-N rule. Percentages with the small-N badge are computed from fewer than 5 records on at least
one side of the ratio. Use the absolute "prior → latest" pair on the
same card; the percentage is mostly noise.
Tier: T1 T2 T3 T4 T5
Section: All sections (912 records) AI sandbox & runtime environments Agent IDEs & coding harnesses Agent frameworks (no first-party memory layer) Browser-agent memory Claude Code memory mechanisms Coding-agent memory Computer-use & desktop agents Dedicated memory layers Embedding & reranker services Enterprise-search adjacencies Evaluation & observability platforms File-backed / editor paradigms Foundation models (substrate reference) Framework-embedded memory Inference platforms & gateways Knowledge-graph platforms Memory benchmarks & evaluation Memory governance, privacy & safety Memory observability & monitoring Multi-agent orchestration platforms Personal AI / PKM / lifelogging memory Platform-provider memory Recent method papers — theorized, no distinct product Research / specialised systems Retrieval-as-memory hybrids Robotics foundation models & agent stacks Search platforms (non-memory) Theoretical / informal — ideas without a paper Training infrastructure Use-case-specific agent harnesses Vector-database infrastructure Vertical / domain-specific AI memory Voice agent platforms Voice-first / wearable AI memory Reset
First appearance by term The earliest year each tracked term shows up in any record's desc or claims within the current filter.
"Newcomers" arrived in 2024 or later — they are the
vocabulary candidates most worth watching.
Newcomers (0) · first seen ≥ 2024 Established (27) · first seen < 2024 1984 1995 2001 2005 2005 2005 2005 2005 2011 2016 2017 2017 2017 2018 2018 2019 2019 2019 2019 2020 2020 2020 2021 2021 2022 2023 2023 Total records 27 YoY (absolute) 12 → 6 (2025 → 2026) YoY (percent) -50% First seen 2017 Latest active 2026-Q2 Top mentions: Anthropic Claude (foundation models) ,LinkedIn Cognitive Memory Agent ,Abridge
Total records 16 YoY (absolute) 8 → 1 (2025 → 2026) YoY (percent) -88% First seen 2017 Latest active 2026-Q2 Top mentions: Botpress LLMz ,Lovable ,OpenAI Agents SDK Memory
Total records 13 YoY (absolute) 5 → 3 (2025 → 2026) YoY (percent) -40% First seen 2020 Latest active 2026-Q2 Top mentions: Anthropic Claude (foundation models) ,OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) ,ICLR 2026 MemAgents Workshop
Total records 45 YoY (absolute) 15 → 7 (2025 → 2026) YoY (percent) -53% First seen 1984 Latest active 2026-Q2 Top mentions: Cline (framework) ,Claude Code (Anthropic) ,GitHub Copilot (Agent Mode)
Total records 7 YoY (absolute) 2 → 1 (2025 → 2026) YoY (percent) -50% small-N First seen 2023 Latest active 2026-Q1 Top mentions: O-Mem ,ELDER ,Voyager
Total records 63 YoY (absolute) 18 → 3 (2025 → 2026) YoY (percent) -83% First seen 2005 Latest active 2026-Q2 Top mentions: Haystack (framework) ,OpenAI Embeddings ,Coveo
Term co-occurrence (top 15) How often two terms appear in the same record's desc + claims. Darker = more co-mentions. Edges with count
≥ 4 form the cluster signal. Click a row label to
drill into that term.
retrieval graph RAG vector semantic agentic persistent MCP episodic hierarchical attention working procedural KV parametric retrieval retrieval ∩ retrieval = 101 (self) 101 retrieval ∩ graph = 17 17 retrieval ∩ RAG = 26 26 retrieval ∩ vector = 17 17 retrieval ∩ semantic = 16 16 retrieval ∩ agentic = 4 4 retrieval ∩ persistent = 2 2 retrieval ∩ MCP = 6 6 retrieval ∩ episodic = 7 7 retrieval ∩ hierarchical = 4 4 retrieval ∩ attention = 4 4 retrieval ∩ working = 1 1 retrieval ∩ procedural = 1 1 retrieval ∩ KV = 4 4 retrieval ∩ parametric = 2 2 graph graph ∩ retrieval = 17 17 graph ∩ graph = 72 (self) 72 graph ∩ RAG = 11 11 graph ∩ vector = 16 16 graph ∩ semantic = 10 10 graph ∩ agentic = 6 6 graph ∩ persistent = 3 3 graph ∩ MCP = 8 8 graph ∩ episodic = 5 5 graph ∩ hierarchical = 5 5 graph ∩ attention = 0 graph ∩ working = 0 graph ∩ procedural = 0 graph ∩ KV = 2 2 graph ∩ parametric = 0 RAG RAG ∩ retrieval = 26 26 RAG ∩ graph = 11 11 RAG ∩ RAG = 64 (self) 64 RAG ∩ vector = 9 9 RAG ∩ semantic = 10 10 RAG ∩ agentic = 7 7 RAG ∩ persistent = 1 1 RAG ∩ MCP = 2 2 RAG ∩ episodic = 4 4 RAG ∩ hierarchical = 2 2 RAG ∩ attention = 2 2 RAG ∩ working = 2 2 RAG ∩ procedural = 1 1 RAG ∩ KV = 3 3 RAG ∩ parametric = 1 1 vector vector ∩ retrieval = 17 17 vector ∩ graph = 16 16 vector ∩ RAG = 9 9 vector ∩ vector = 56 (self) 56 vector ∩ semantic = 7 7 vector ∩ agentic = 4 4 vector ∩ persistent = 2 2 vector ∩ MCP = 5 5 vector ∩ episodic = 2 2 vector ∩ hierarchical = 1 1 vector ∩ attention = 0 vector ∩ working = 3 3 vector ∩ procedural = 1 1 vector ∩ KV = 1 1 vector ∩ parametric = 1 1 semantic semantic ∩ retrieval = 16 16 semantic ∩ graph = 10 10 semantic ∩ RAG = 10 10 semantic ∩ vector = 7 7 semantic ∩ semantic = 53 (self) 53 semantic ∩ agentic = 2 2 semantic ∩ persistent = 5 5 semantic ∩ MCP = 2 2 semantic ∩ episodic = 12 12 semantic ∩ hierarchical = 3 3 semantic ∩ attention = 1 1 semantic ∩ working = 3 3 semantic ∩ procedural = 3 3 semantic ∩ KV = 0 semantic ∩ parametric = 1 1 agentic agentic ∩ retrieval = 4 4 agentic ∩ graph = 6 6 agentic ∩ RAG = 7 7 agentic ∩ vector = 4 4 agentic ∩ semantic = 2 2 agentic ∩ agentic = 46 (self) 46 agentic ∩ persistent = 4 4 agentic ∩ MCP = 3 3 agentic ∩ episodic = 0 agentic ∩ hierarchical = 2 2 agentic ∩ attention = 0 agentic ∩ working = 1 1 agentic ∩ procedural = 0 agentic ∩ KV = 1 1 agentic ∩ parametric = 0 persistent persistent ∩ retrieval = 2 2 persistent ∩ graph = 3 3 persistent ∩ RAG = 1 1 persistent ∩ vector = 2 2 persistent ∩ semantic = 5 5 persistent ∩ agentic = 4 4 persistent ∩ persistent = 40 (self) 40 persistent ∩ MCP = 2 2 persistent ∩ episodic = 0 persistent ∩ hierarchical = 0 persistent ∩ attention = 0 persistent ∩ working = 1 1 persistent ∩ procedural = 0 persistent ∩ KV = 1 1 persistent ∩ parametric = 0 MCP MCP ∩ retrieval = 6 6 MCP ∩ graph = 8 8 MCP ∩ RAG = 2 2 MCP ∩ vector = 5 5 MCP ∩ semantic = 2 2 MCP ∩ agentic = 3 3 MCP ∩ persistent = 2 2 MCP ∩ MCP = 37 (self) 37 MCP ∩ episodic = 1 1 MCP ∩ hierarchical = 0 MCP ∩ attention = 0 MCP ∩ working = 0 MCP ∩ procedural = 0 MCP ∩ KV = 0 MCP ∩ parametric = 0 episodic episodic ∩ retrieval = 7 7 episodic ∩ graph = 5 5 episodic ∩ RAG = 4 4 episodic ∩ vector = 2 2 episodic ∩ semantic = 12 12 episodic ∩ agentic = 0 episodic ∩ persistent = 0 episodic ∩ MCP = 1 1 episodic ∩ episodic = 27 (self) 27 episodic ∩ hierarchical = 1 1 episodic ∩ attention = 0 episodic ∩ working = 3 3 episodic ∩ procedural = 3 3 episodic ∩ KV = 1 1 episodic ∩ parametric = 3 3 hierarchical hierarchical ∩ retrieval = 4 4 hierarchical ∩ graph = 5 5 hierarchical ∩ RAG = 2 2 hierarchical ∩ vector = 1 1 hierarchical ∩ semantic = 3 3 hierarchical ∩ agentic = 2 2 hierarchical ∩ persistent = 0 hierarchical ∩ MCP = 0 hierarchical ∩ episodic = 1 1 hierarchical ∩ hierarchical = 25 (self) 25 hierarchical ∩ attention = 0 hierarchical ∩ working = 1 1 hierarchical ∩ procedural = 0 hierarchical ∩ KV = 0 hierarchical ∩ parametric = 1 1 attention attention ∩ retrieval = 4 4 attention ∩ graph = 0 attention ∩ RAG = 2 2 attention ∩ vector = 0 attention ∩ semantic = 1 1 attention ∩ agentic = 0 attention ∩ persistent = 0 attention ∩ MCP = 0 attention ∩ episodic = 0 attention ∩ hierarchical = 0 attention ∩ attention = 23 (self) 23 attention ∩ working = 1 1 attention ∩ procedural = 0 attention ∩ KV = 4 4 attention ∩ parametric = 0 working working ∩ retrieval = 1 1 working ∩ graph = 0 working ∩ RAG = 2 2 working ∩ vector = 3 3 working ∩ semantic = 3 3 working ∩ agentic = 1 1 working ∩ persistent = 1 1 working ∩ MCP = 0 working ∩ episodic = 3 3 working ∩ hierarchical = 1 1 working ∩ attention = 1 1 working ∩ working = 16 (self) 16 working ∩ procedural = 2 2 working ∩ KV = 2 2 working ∩ parametric = 1 1 procedural procedural ∩ retrieval = 1 1 procedural ∩ graph = 0 procedural ∩ RAG = 1 1 procedural ∩ vector = 1 1 procedural ∩ semantic = 3 3 procedural ∩ agentic = 0 procedural ∩ persistent = 0 procedural ∩ MCP = 0 procedural ∩ episodic = 3 3 procedural ∩ hierarchical = 0 procedural ∩ attention = 0 procedural ∩ working = 2 2 procedural ∩ procedural = 15 (self) 15 procedural ∩ KV = 0 procedural ∩ parametric = 0 KV KV ∩ retrieval = 4 4 KV ∩ graph = 2 2 KV ∩ RAG = 3 3 KV ∩ vector = 1 1 KV ∩ semantic = 0 KV ∩ agentic = 1 1 KV ∩ persistent = 1 1 KV ∩ MCP = 0 KV ∩ episodic = 1 1 KV ∩ hierarchical = 0 KV ∩ attention = 4 4 KV ∩ working = 2 2 KV ∩ procedural = 0 KV ∩ KV = 14 (self) 14 KV ∩ parametric = 0 parametric parametric ∩ retrieval = 2 2 parametric ∩ graph = 0 parametric ∩ RAG = 1 1 parametric ∩ vector = 1 1 parametric ∩ semantic = 1 1 parametric ∩ agentic = 0 parametric ∩ persistent = 0 parametric ∩ MCP = 0 parametric ∩ episodic = 3 3 parametric ∩ hierarchical = 1 1 parametric ∩ attention = 0 parametric ∩ working = 1 1 parametric ∩ procedural = 0 parametric ∩ KV = 0 parametric ∩ parametric = 13 (self) 13
Threshold ≥ 4 shown with orange border.
Max off-diagonal: 26.
Stable vs novel vocabulary per year For each year, the share of new records whose desc + claims mention at least one of our 27
curated terms (stable) vs records using none of them (novel /
drifting). A rising novel share = the field's vocabulary is
outrunning our curated list. Bar width is the record count for the
year (normalised to the largest); rising novel slice = vocabulary
churn.
Year Records Stable Novel Stable share Mix 1984 1 1 0 100% 1995 2 1 1 50% 1999 1 0 1 0% 2001 2 1 1 50% 2003 1 1 0 100% 2005 2 2 0 100% 2006 1 0 1 0% 2007 3 1 2 33% 2008 2 2 0 100% 2009 3 0 3 0% 2010 3 2 1 67% 2011 3 1 2 33% 2012 7 5 2 71% 2013 5 3 2 60% 2014 11 3 8 27% 2015 15 2 13 13% 2016 11 4 7 36% 2017 19 4 15 21% 2018 31 10 21 32% 2019 33 12 21 36% 2020 32 10 22 31% 2021 38 13 25 34% 2022 62 23 39 37% 2023 182 70 112 38% 2024 212 97 115 46% 2025 168 104 64 62% 2026 43 32 11 74%