Six panels. The first five combine funding cadence, release recency, GitHub stars, mindshare/citations, inbound integration edges, and lineage membership into a heuristic velocity story per record — confidence labels are surfaced everywhere so a reader can weigh each callout. The sixth fits a BIMATEM-style logistic S-curve to every row with enough temporal signal and reads pre-growth / growth / saturation / decline straight off the inflection — quantitative complement to the heuristic buckets, and the math the Gartner Hype Cycle waves at without ever publishing.
Every record bucketed. Click a counter above to filter to one bucket; sort by clicking column headers. Each row drills to the main catalog table.
| Section | |||||
|---|---|---|---|---|---|
| π0.5 (Physical Intelligence) | T1 | Vertical / domain-specific AI memory | Accelerating | high | funded 6mo ago (Series B · $1.07B) · 13,300 GH stars |
| Abridge | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 1mo ago (Series E · $616M) |
| ACE — Agentic Context Engineering | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 1387 |
| Acontext (memodb) | T3 | Dedicated memory layers | Accelerating | medium | mindshare 3696 |
| Acuvity (now Proofpoint) | T1 | Memory governance, privacy & safety | Accelerating | medium | funded 3mo ago (Seed · $9M) |
| Agent QA | T3 | Use-case-specific agent harnesses | Accelerating | medium | mindshare 897 |
| Agent Router (formerly Envoy AI Gateway) | T3 | Inference platforms & gateways | Accelerating | medium | mindshare 2314 |
| Agent S | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 147 |
| Agent Workflow Memory | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 133 |
| Agent2Agent Protocol (A2A) | T2 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 25,500 GH stars |
| AgentBench | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 3798 |
| agentgateway | T3 | Inference platforms & gateways | Accelerating | medium | mindshare 5769 |
| AGENTS.md | T1 | File-backed / editor paradigms | Accelerating | medium | 25,000 GH stars |
| Agility Robotics | T2 | Robotics foundation models & agent stacks | Accelerating | medium | funded -6mo ago ($2.5B) |
| AGiXT Adaptive Memory | T2 | Framework-embedded memory | Accelerating | medium | mindshare 3781 |
| Agno (Phidata) Memory | T2 | Framework-embedded memory | Accelerating | medium | 42,800 GH stars |
| Aider | T2 | Coding-agent memory | Accelerating | medium | 41,000 GH stars |
| Aider (harness) | T2 | Agent IDEs & coding harnesses | Accelerating | medium | 49,700 GH stars |
| Alita | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 92 |
| AlphaEdit | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 191 |
| Anthropic Claude (foundation models) | T1 | Foundation models (substrate reference) | Accelerating | medium | funded -5mo ago (Series H · $65B) |
| Anthropic Claude Memory | T1 | Platform-provider memory | Accelerating | medium | funded 0mo ago (Series H · $65B) |
| Anthropic Computer Use | T1 | Computer-use & desktop agents | Accelerating | medium | funded -5mo ago (Series H · $65B) |
| Anthropic Managed Agents — Memory | T3 | Platform-provider memory | Accelerating | medium | funded -5mo ago (Series H · $65B) |
| Anyscale | T1 | Inference platforms & gateways | Accelerating | medium | 44,100 GH stars |
| ArangoDB | T1 | Knowledge-graph platforms | Accelerating | medium | 14,300 GH stars |
| Arize Phoenix | T1 | Evaluation & observability platforms | Accelerating | medium | 12,000 GH stars |
| Arthur Engine | T3 | Guardrails & AI security | Accelerating | medium | mindshare 93 |
| Augment Code | T1 | Coding-agent memory | Accelerating | medium | funded -5mo ago (Acquired · $252M) |
| AutoGen Memory | T2 | Framework-embedded memory | Accelerating | medium | 60,400 GH stars |
| AutoGLM (Zhipu AI) | T2 | Dedicated memory layers | Accelerating | high | funded 4mo ago (IPO · $4.35B) · 26,200 GH stars |
| AutoGPT | T2 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 187,700 GH stars |
| Axolotl | T2 | Training infrastructure | Accelerating | medium | 12,600 GH stars |
| BABILong | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 196 |
| BabyAGI | T2 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 22,300 GH stars |
| Backboard | T2 | Dedicated memory layers | Accelerating | medium | funded 4mo ago (pre-seed · $3M) |
| Basic Memory | T3 | File-backed / editor paradigms | Accelerating | medium | mindshare 4322 |
| BEAM (Beyond a Million Tokens) + LIGHT | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 164 |
| Bee | T1 | Voice-first / wearable AI memory | Accelerating | medium | funded 10mo ago (seed · $7M) |
| BeeAI Framework | T3 | Agent frameworks (no first-party memory layer) | Accelerating | medium | mindshare 3443 |
| τ-bench (tau-bench / τ²-bench) | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 2354 |
| BGE (BAAI) | T2 | Embedding & reranker services | Accelerating | medium | 12,200 GH stars |
| BGE-M3 | T4 | Retrieval-as-memory hybrids | Accelerating | medium | 12,200 GH stars |
| Bifrost (Maxim AI) | T3 | Inference platforms & gateways | Accelerating | medium | mindshare 9198 |
| Bolt.new (StackBlitz) | T2 | Coding-agent memory | Accelerating | medium | 16,600 GH stars |
| Bolt.new (StackBlitz) — harness | T1 | Agent IDEs & coding harnesses | Accelerating | medium | 16,600 GH stars |
| Botpress LLMz | T1 | Framework-embedded memory | Accelerating | medium | 14,900 GH stars |
| BrowseComp | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 4686 |
| Browser Use | T1 | Computer-use & desktop agents | Accelerating | medium | 119,000 GH stars |
| Browserbase | T1 | Browser-agent memory | Accelerating | medium | funded 11mo ago (Series B · $67.5M) |
| Browserless.io | T2 | AI sandbox & runtime environments | Accelerating | medium | 13,800 GH stars |
| Buffer of Thoughts | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 95 |
| Bytebot | T3 | Computer-use & desktop agents | Accelerating | medium | 11,100 GH stars |
| Calypso AI (now F5 AI Guardrails) | T2 | Use-case-specific agent harnesses | Accelerating | medium | funded 8mo ago (acquired · $40M) |
| Cellon (Memory.inc) | T2 | Dedicated memory layers | Accelerating | medium | funded 1mo ago (Seed) |
| Character.ai | T1 | Vertical / domain-specific AI memory | Accelerating | high | funded -5mo ago ($2.7B) · mindshare 84 |
| Charm Crush | T3 | Agent IDEs & coding harnesses | Accelerating | medium | 27,900 GH stars |
| ChatGPT Study Mode | T3 | Vertical / domain-specific AI memory | Accelerating | medium | funded 1mo ago ($122B) |
| Chroma | T1 | Vector-database infrastructure | Accelerating | high | funded -5mo ago (Acquired · $18M) · 29,100 GH stars |
| CL of LLMs Survey | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 237 |
| Claude Agent SDK | T3 | Agent frameworks (no first-party memory layer) | Accelerating | medium | mindshare 8347 |
| Claude Code (Anthropic) | T1 | Agent IDEs & coding harnesses | Accelerating | high | funded -5mo ago (Series H · $65B) · 153,000 GH stars |
| Claude for Chrome | T1 | Computer-use & desktop agents | Accelerating | medium | funded -5mo ago (Series H · $65B) |
| claude-mem | T1 | Claude Code memory mechanisms | Accelerating | high | 104,000 GH stars · mindshare 72 |
| Cline (framework) | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 67,600 GH stars |
| Cline (harness) | T2 | Agent IDEs & coding harnesses | Accelerating | medium | 67,600 GH stars |
| Clio Duo / Manage AI | T2 | Vertical / domain-specific AI memory | Accelerating | medium | funded 6mo ago (Series G · $1.7B) |
| codebase-memory-mcp (DeusData) | T1 | Claude Code memory mechanisms | Accelerating | medium | 43,300 GH stars |
| Coder.com | T2 | AI sandbox & runtime environments | Accelerating | medium | 14,100 GH stars |
| Cognee | T1 | Dedicated memory layers | Accelerating | high | funded 3mo ago (Seed · $10M) · 31,200 GH stars |
| Cognee MCP | T2 | Claude Code memory mechanisms | Accelerating | medium | 31,200 GH stars |
| Cognigy (now NiCE Cognigy) | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 8mo ago (acquired · $100M) |
| Cognition Devin v2 / Spec Mode | T1 | Agent IDEs & coding harnesses | Accelerating | medium | funded -4mo ago (Series E · $2B) |
| ColossalAI | T2 | Training infrastructure | Accelerating | medium | 41,400 GH stars |
| Comet ML | T2 | Training infrastructure | Accelerating | medium | 23,000 GH stars |
| Composio | T2 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 30,000 GH stars |
| Compressive Transformer | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 832 |
| Confident AI / DeepEval | T2 | Evaluation & observability platforms | Accelerating | medium | 19,100 GH stars |
| Continue.dev .continue/rules | T2 | File-backed / editor paradigms | Accelerating | high | funded -1mo ago (acquired) · 35,700 GH stars |
| Continue.dev | T2 | Agent IDEs & coding harnesses | Accelerating | high | funded -1mo ago (Acquired · $7.5M) · 35,700 GH stars |
| Continue.dev Memory MCP | T2 | Coding-agent memory | Accelerating | medium | funded -1mo ago (acquired · $5.6M) |
| CORE (RedPlanet) | T3 | Personal AI / PKM / lifelogging memory | Accelerating | medium | mindshare 2003 |
| CREATOR | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 70 |
| CrewAI | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 58,600 GH stars |
| CrewAI Enterprise (now CrewAI AMP) | T2 | Multi-agent orchestration platforms | Accelerating | medium | 58,600 GH stars |
| CrowdStrike Falcon Guardian (formerly Pangea AI Guard) | T3 | Guardrails & AI security | Accelerating | medium | funded 8mo ago (acquired) |
| Cursor (IDE) | T1 | Agent IDEs & coding harnesses | Accelerating | medium | funded -3mo ago (Acquired · $900M) |
| Dagster | T1 | Training infrastructure | Accelerating | high | funded -2mo ago (Acquired · $47M) · 16,300 GH stars |
| Dapr Agents | T3 | Durable execution & workflow engines | Accelerating | medium | mindshare 768 |
| Darwin Gödel Machine | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 84 |
| Daytona | T2 | AI sandbox & runtime environments | Accelerating | medium | 71,000 GH stars |
| DBOS | T3 | Durable execution & workflow engines | Accelerating | medium | mindshare 1654 |
| Decagon | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 4mo ago (Series D · $250M) |
| DeepSeek R1 / V3 family | T1 | Foundation models (substrate reference) | Accelerating | medium | 94,000 GH stars |
| DeepSpeed | T1 | Training infrastructure | Accelerating | medium | 43,300 GH stars |
| Dgraph | T2 | Knowledge-graph platforms | Accelerating | medium | 21,800 GH stars |
| Dia (Atlassian) | T1 | Browser-agent memory | Accelerating | medium | funded 7mo ago (acquisition · $128M) |
| DIAMOND | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 209 |
| Differentiable Search Index (DSI) | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 405 |
| Docker MCP Catalog, Toolkit & Gateway | T3 | MCP gateways & registries | Accelerating | medium | mindshare 1616 |
| DreamerV3 | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 1032 |
| DSPy History | T3 | Framework-embedded memory | Accelerating | medium | 36,700 GH stars |
| DVC | T1 | Training infrastructure | Accelerating | medium | 15,900 GH stars |
| e2b | T1 | AI sandbox & runtime environments | Accelerating | medium | 15,000 GH stars |
| Elasticsearch | T1 | Vector-database infrastructure | Accelerating | medium | 78,000 GH stars |
| Eliza (ai16z) | T2 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 20,000 GH stars |
| Enkrypt AI | T1 | Memory governance, privacy & safety | Accelerating | medium | funded -3mo ago (Acquired · $2.35M) |
| Era Computer | T5 | Voice-first / wearable AI memory | Accelerating | medium | funded 1mo ago (Seed · $11M) |
| Eve — Shared Firm Knowledge + Auditor | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 7mo ago (Series B · $103M) |
| EWC (Elastic Weight Consolidation) | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 9657 |
| ExpeL | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 481 |
| FalkorDB | T2 | Knowledge-graph platforms | Accelerating | medium | 10,500 GH stars |
| Figure Helix | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 8mo ago (Series C · $1B) |
| Filevine LOIS | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 8mo ago ($400M) |
| Fincon | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 124 |
| Flowise Memory | T1 | Framework-embedded memory | Accelerating | high | funded 9mo ago (Seed) · 55,400 GH stars |
| From Human Memory to AI Memory (survey) | T4 | Theoretical / informal — ideas without a paper | Accelerating | medium | mindshare 73 |
| FSDP (PyTorch) | T1 | Training infrastructure | Accelerating | medium | 105,000 GH stars |
| Galileo | T1 | Evaluation & observability platforms | Accelerating | medium | funded 0mo ago (Series B · $45M) |
| Galileo (galileo.ai) (now Splunk Agent Observability) | T1 | Memory observability & monitoring | Accelerating | medium | funded 0mo ago (Acquired · $68M) |
| Generative Agents | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 3905 |
| Genie | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 513 |
| Genkit | T3 | Agent frameworks (no first-party memory layer) | Accelerating | medium | mindshare 6549 |
| GitHub Copilot (Agent Mode) | T1 | Agent IDEs & coding harnesses | Accelerating | medium | mindshare 100 |
| Gitpod AI features (now Ona) | T2 | AI sandbox & runtime environments | Accelerating | high | funded -3mo ago (Acquired · $25M) · 14,000 GH stars |
| Glean | T1 | Enterprise-search adjacencies | Accelerating | high | funded 11mo ago (Series F · $765M) · mindshare 660 |
| Golem | T3 | Durable execution & workflow engines | Accelerating | medium | mindshare 1523 |
| Google ADK Memory | T2 | Framework-embedded memory | Accelerating | medium | 21,100 GH stars |
| Google DeepMind Gemini Robotics | T1 | Robotics foundation models & agent stacks | Accelerating | medium | newest member of "Robotics foundation-model thread" lineage |
| Goose (Block) | T2 | Agent IDEs & coding harnesses | Accelerating | medium | 55,700 GH stars |
| GPT Engineer | T3 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 55,000 GH stars |
| Graphiti MCP Server (Zep) | T1 | Claude Code memory mechanisms | Accelerating | medium | 31,900 GH stars |
| GraphRAG (Microsoft) | T1 | Retrieval-as-memory hybrids | Accelerating | medium | 35,100 GH stars |
| Grok Memory (xAI) | T1 | Platform-provider memory | Accelerating | medium | funded 3mo ago (Series E · $20B) |
| Guardrails AI | T3 | Guardrails & AI security | Accelerating | high | funded -4mo ago (Acquired) · mindshare 7568 |
| H2O | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 654 |
| HaluMem | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 178 |
| Harvey Memory | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded -4mo ago ($550M) |
| Hatchet | T3 | Durable execution & workflow engines | Accelerating | medium | mindshare 8198 |
| Haystack (framework) | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 27,000 GH stars |
| Haystack Memory (deepset) | T3 | Framework-embedded memory | Accelerating | medium | 26,200 GH stars |
| Heidi Health | T2 | Vertical / domain-specific AI memory | Accelerating | medium | funded -4mo ago (Series C · $436.6M) |
| HELMET | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 96 |
| HiddenLayer AISec Platform 2.0 | T1 | Memory governance, privacy & safety | Accelerating | medium | funded -4mo ago (Series B · $156M) |
| Hindsight MCP | T2 | Claude Code memory mechanisms | Accelerating | medium | 44,600 GH stars |
| Hindsight (Vectorize) | T1 | Dedicated memory layers | Accelerating | medium | 44,600 GH stars |
| Hippocratic AI Polaris | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 6mo ago (Series C · $444M) |
| Honcho (Plastic Labs) | T3 | Dedicated memory layers | Accelerating | medium | mindshare 7730 |
| Hugging Face Datasets | T1 | Training infrastructure | Accelerating | medium | 22,100 GH stars |
| Hugging Face LeRobot | T1 | Robotics foundation models & agent stacks | Accelerating | medium | 28,500 GH stars |
| Hyperspell | T2 | Dedicated memory layers | Accelerating | medium | funded 8mo ago (Seed · $2M) |
| I-JEPA | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 773 |
| IBM ContextForge MCP Gateway | T3 | MCP gateways & registries | Accelerating | medium | mindshare 4720 |
| Infini-attention | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 192 |
| InfiniteBench (∞Bench) | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 330 |
| Inspect AI | T3 | Evaluation & observability platforms | Accelerating | medium | mindshare 3128 |
| Interloom | T1 | Dedicated memory layers | Accelerating | medium | funded 2mo ago (Seed · $16M) |
| Invariant Guardrails | T3 | Guardrails & AI security | Accelerating | high | funded 11mo ago (acquired) · mindshare 475 |
| Jamba (AI21 Labs) | T1 | Foundation models (substrate reference) | Accelerating | medium | funded 12mo ago (Series D · $608M) |
| JARVIS-1 | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 179 |
| Jina AI Embeddings | T1 | Embedding & reranker services | Accelerating | medium | funded 7mo ago (Acquired · $37.5M) |
| K Health | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded -3mo ago (Series F · $58.9M) |
| Kiro | T1 | Agent IDEs & coding harnesses | Accelerating | medium | newest member of "Specs-as-memory thread" lineage |
| Label Studio (HumanSignal) | T1 | Training infrastructure | Accelerating | medium | 28,600 GH stars |
| Labelbox | T1 | Training infrastructure | Accelerating | medium | funded -5mo ago (Series E · $100M) |
| Lakera Guard / Lakera Red | T1 | Memory governance, privacy & safety | Accelerating | medium | funded 6mo ago (acquired · $300M) |
| Laminar | T3 | Evaluation & observability platforms | Accelerating | medium | mindshare 3460 |
| LanceDB | T1 | Vector-database infrastructure | Accelerating | high | funded 11mo ago (Series A · $38M) · 11,300 GH stars |
| LangChain (framework) | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 149,000 GH stars |
| Langflow Memory | T1 | Framework-embedded memory | Accelerating | medium | 156,000 GH stars |
| Langfuse | T1 | Memory observability & monitoring | Accelerating | high | funded 4mo ago (Acquired) · 36,100 GH stars |
| LangGraph (orchestration) | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 44,000 GH stars |
| LangGraph Persistence | T2 | Framework-embedded memory | Accelerating | high | funded 7mo ago (Series B · $125M) · 43,000 GH stars |
| LangMem (LangChain) | T2 | Framework-embedded memory | Accelerating | medium | funded 7mo ago (Series B · $160M) |
| LazyGraphRAG | T3 | Retrieval-as-memory hybrids | Accelerating | medium | 35,100 GH stars |
| Letta Code | T3 | Coding-agent memory | Accelerating | medium | mindshare 3683 |
| LightRAG | T3 | Retrieval-as-memory hybrids | Accelerating | medium | 39,300 GH stars |
| LightRAG | T3 | Retrieval-as-memory hybrids | Accelerating | medium | 39,300 GH stars |
| Limitless | T1 | Voice-first / wearable AI memory | Accelerating | medium | funded 5mo ago (Acquired · $34.3M) |
| LiteLLM | T1 | Inference platforms & gateways | Accelerating | medium | 62,100 GH stars |
| LiveKit Agents | T1 | Voice agent platforms | Accelerating | medium | 13,000 GH stars |
| LLaMA-Factory | T2 | Training infrastructure | Accelerating | medium | 75,700 GH stars |
| LlamaIndex (framework) | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 53,000 GH stars |
| LlamaIndex Memory | T2 | Framework-embedded memory | Accelerating | medium | 52,600 GH stars |
| local-memory-mcp | T2 | Claude Code memory mechanisms | Accelerating | medium | newest member of "Files-as-memory thread" lineage |
| LoCoMo | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 395 |
| Logseq | T2 | Personal AI / PKM / lifelogging memory | Accelerating | high | 45,000 GH stars · mindshare 108 |
| LongBench | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 1200 |
| LongMem | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 172 |
| LongMemEval | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 240 |
| LoRA | T3 | Training infrastructure | Accelerating | medium | 13,900 GH stars |
| Lovable | T2 | Coding-agent memory | Accelerating | medium | funded -3mo ago (Series C · $400M) |
| Lovable.dev — harness | T1 | Agent IDEs & coding harnesses | Accelerating | medium | funded -3mo ago (Series C · $400M) |
| Lunar.dev MCPX | T3 | MCP gateways & registries | Accelerating | high | funded -2mo ago (Acquired) · mindshare 506 |
| m_flow (M-Flow) | T3 | Retrieval-as-memory hybrids | Accelerating | medium | mindshare 4512 |
| Magnetic-One | T3 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 62,000 GH stars |
| Mamba-2 / SSD | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | newest member of "State-space models" lineage |
| Mastra | T2 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 29,000 GH stars |
| Mastra Memory | T2 | Framework-embedded memory | Accelerating | high | funded 4mo ago (Series A · $35M) · 27,100 GH stars |
| mcp-memory (Puliczek) | T2 | Claude Code memory mechanisms | Accelerating | medium | mindshare 142 |
| MCP Servers (Reference) | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 92,000 GH stars |
| Megatron-Core | T2 | Training infrastructure | Accelerating | medium | 18,300 GH stars |
| Megatron-LM | T1 | Training infrastructure | Accelerating | medium | 18,300 GH stars |
| Meilisearch | T2 | Enterprise-search adjacencies | Accelerating | medium | 59,600 GH stars |
| Mem0 | T1 | Dedicated memory layers | Accelerating | high | funded 7mo ago (Series A · $24M) · 67,300 GH stars |
| Mem0 MCP (official) | T1 | Claude Code memory mechanisms | Accelerating | medium | mindshare 650 |
| Mem0 Security / OpenMemory | T1 | Memory governance, privacy & safety | Accelerating | medium | 67,300 GH stars |
| MEM1 | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 107 |
| MemAgent | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 123 |
| Memanto (Moorcheh) | T3 | Dedicated memory layers | Accelerating | medium | mindshare 2337 |
| Memformer | T3 | Recent method papers — theorized, no distinct product | Accelerating | medium | mindshare 77 |
| MemoraX AI | T2 | Dedicated memory layers | Accelerating | medium | funded 2mo ago (Seed · $10M) |
| Memori | T2 | Dedicated memory layers | Accelerating | medium | 17,100 GH stars |
| Memori (GibsonAI) (now Memori Labs) | T3 | Dedicated memory layers | Accelerating | medium | 17,100 GH stars |
| Memories.ai LVMM 2.0 / Project LUCI | T2 | Voice-first / wearable AI memory | Accelerating | medium | funded -1mo ago (seed · $16M) |
| memory-lancedb-pro (OpenClaw plugin) | T3 | Framework-embedded memory | Accelerating | medium | mindshare 4452 |
| MemoryBear | T3 | Dedicated memory layers | Accelerating | medium | mindshare 8260 |
| memsearch (Zilliz) | T3 | Coding-agent memory | Accelerating | medium | mindshare 2775 |
| Memvid | T2 | Dedicated memory layers | Accelerating | medium | 16,600 GH stars |
| Meta Llama 4 family | T1 | Foundation models (substrate reference) | Accelerating | medium | 58,000 GH stars |
| Meta Llama Protections (Llama Guard, Prompt Guard, LlamaFirewall) | T3 | Guardrails & AI security | Accelerating | medium | mindshare 4450 |
| MetaGPT | T2 | Agent IDEs & coding harnesses | Accelerating | medium | 71,000 GH stars |
| MIA — Memory Intelligence Agent | T4 | Recent method papers — theorized, no distinct product | Accelerating | medium | newest member of "EWC (Elastic Weight Consolidation) family" lineage |
| Microsoft AutoGen Studio | T2 | Multi-agent orchestration platforms | Accelerating | medium | 62,000 GH stars |
| Microsoft MCP Gateway | T3 | MCP gateways & registries | Accelerating | medium | mindshare 899 |
| Microsoft Semantic Kernel | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 29,000 GH stars |
| Milvus | T1 | Vector-database infrastructure | Accelerating | medium | 46,000 GH stars |
| Mind2Web | T3 | Memory benchmarks & evaluation | Accelerating | medium | mindshare 1036 |
| Mistral Agents API | T1 | Platform-provider memory | Accelerating | medium | funded -4mo ago (Series D) |
| Mistral Le Chat — Memories (now Vibe) | T1 | Platform-provider memory | Accelerating | medium | funded -4mo ago (Series D) |
| MLflow | T1 | Training infrastructure | Accelerating | medium | 28,600 GH stars |
| n8n (AI agent flows) | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 209,000 GH stars |
| n8n AI Agent Memory | T1 | Framework-embedded memory | Accelerating | medium | 208,000 GH stars |
| Nabla + Navina | T1 | Vertical / domain-specific AI memory | Accelerating | medium | funded 11mo ago (Series C · $120M) |
| NebulaGraph | T2 | Knowledge-graph platforms | Accelerating | medium | 12,400 GH stars |
| Neo4j | T1 | Knowledge-graph platforms | Accelerating | medium | 17,200 GH stars |
| Neo4j Agent Memory (Neo4j Labs) | T3 | Dedicated memory layers | Accelerating | medium | mindshare 592 |
| NeoCognition | T1 | Dedicated memory layers | Accelerating | medium | funded 1mo ago (Seed · $40M) |
| NVIDIA NeMo Guardrails | T3 | Guardrails & AI security | Accelerating | medium | mindshare 7394 |
| Nyne | T2 | Dedicated memory layers | Accelerating | medium | funded 2mo ago (Seed · $5.3M) |
| Obot | T3 | MCP gateways & registries | Accelerating | medium | mindshare 1136 |
| Official MCP Memory server | T1 | Claude Code memory mechanisms | Accelerating | medium | 90,000 GH stars |
| Official MCP Registry | T3 | MCP gateways & registries | Accelerating | medium | mindshare 7419 |
| Omi | T1 | Voice-first / wearable AI memory | Accelerating | medium | 13,200 GH stars |
| OpenAI Agents SDK | T1 | Agent frameworks (no first-party memory layer) | Accelerating | medium | 30,000 GH stars |
| OpenAI Agents SDK Memory | T2 | Framework-embedded memory | Accelerating | medium | 28,600 GH stars |
| OpenAI ChatGPT Memory | T1 | Platform-provider memory | Accelerating | medium | funded 1mo ago ($122B) |
| OpenAI Codex CLI | T1 | Agent IDEs & coding harnesses | Accelerating | medium | 128,000 GH stars |
| OpenAI Codex (cloud agent) | T1 | Coding-agent memory | Accelerating | high | funded 1mo ago ($122B) · 128,000 GH stars |
| OpenAI Embeddings | T1 | Embedding & reranker services | Accelerating | medium | 32,000 GH stars |
Showing first 250 of 1002 rows. Narrow the filters above to focus the view.
For records that name a foundation-model substrate (in claims / desc / llm-lock / embedding-model cells), who depends on what. Single-vendor exposure is flagged: if the substrate disappears or restricts access, every member here feels it on the same day.
Predicted M&A categories with recent precedents. Top 3 candidates per category, scored by inbound integrations + tier + funding recency. Speculative — labels match the issue's seed framing, not Anthropic guidance.
Precedent: Voyage → MongoDB (Feb 2025)
Precedent: Quickwit → Datadog (Dec 2024) · OctoAI → NVIDIA (Oct 2024)
Precedent: Pinecone serverless pivot
Combining recent funding rounds, climbing GitHub stars, integration adoptions, and mindshare. Confidence labels reflect how many signals agree. Speculative — calibrated against precedent rather than insider data.
recent funding (-5mo · Series H · $65B) · 153k stars · 27 integrations
recent funding (3mo · Seed · $10M) · 31k stars · 8 integrations
recent funding (9mo · Seed) · 55k stars · 7 integrations
recent funding (4mo · Acquired) · 36k stars · 26 integrations
recent funding (7mo · Series A · $24M) · 67k stars · 24 integrations
recent funding (2mo · Series B · $78M) · 35k stars · 22 integrations
recent funding (3mo · Series E · $20B) · 50k stars · 30 integrations
recent funding (-1mo · Acquired · $7.5M) · 36k stars · 5 integrations
recent funding (7mo · Series B · $125M) · 43k stars · 5 integrations
recent funding (-5mo · Acquired · $18M) · 29k stars · 17 integrations
Low integration count + low citation rate + slow lineage cadence + no/declining funding. The "dead" rows from Panel 1 are excluded here — those are already gone. This list is the slope-down rows that haven't flatlined yet.
Quantitative complement to Panels 1-5. For every row we fit a 3-parameter logistic — y = L / (1 + exp(-k·(t-t0))) — to its cumulative temporal series (citations / dated milestones / OSS stars). The inflection date t0 divides growth from saturation; the fitted k tells us how steep the transition is. This is the math Gartner's Hype Cycle waves at without ever publishing — ours has the math.
204 rows fit · median R² = 0.98 · click a counter to filter the table below.
| Section | Curve | L | Signal | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Mastra Memory | Framework-embedded memory | Saturation | 1.00 | 0.317 | 13626k | inflection passed Q2 2026 | 53 | 6,387,189 downloads · 53mo | |
| Self-Refine | Recent method papers — theorized, no distinct product | Saturation | 1.00 | 0.115 | 12 | inflection passed Q2 2024 | 13 | 11 cites (1 infl) · 4yr | |
| Accelerate (Hugging Face) | Training infrastructure | Saturation | 1.00 | 0.088 | 2.0k | inflection passed Q3 2023 | 35 | 1,956 commits · 35mo | |
| RAFT | Retrieval-as-memory hybrids | Saturation | 1.00 | 0.223 | 470 | inflection passed Q4 2024 | 33 | 470 commits · 33mo | |
| OpenInference | Evaluation & observability platforms | Saturation | 1.00 | 0.099 | 2.6k | inflection passed Q3 2025 | 16 | 1,853 commits · 16mo | |
| ITER-RETGEN | Retrieval-as-memory hybrids | Growth | 1.00 | 0.040 | 10 | saturation expected ~Q2 2027 | 9 | 4 cites (1 infl) · 3yr | |
| Retroformer | Recent method papers — theorized, no distinct product | Growth | 1.00 | 0.040 | 10 | saturation expected ~Q2 2026 | 9 | 4 cites (0 infl) · 3yr | |
| Transformer² | Recent method papers — theorized, no distinct product | Saturation | 1.00 | 0.273 | 17 | inflection passed Q1 2026 | 6 | 13 cites · 13/yr · 1.4yr | |
| Steel.dev | Computer-use & desktop agents | Saturation | 1.00 | 0.271 | 245 | inflection passed Q1 2025 | 18 | 245 commits · 18mo | |
| Generative Agents | Recent method papers — theorized, no distinct product | Saturation | 1.00 | 0.080 | 36 | inflection passed Q3 2024 | 13 | 31 cites (7 infl) · 4yr | |
| V-JEPA 2 | Recent method papers — theorized, no distinct product | Saturation | 1.00 | 0.429 | 316 | inflection passed Q3 2025 | 10 | 316 cites · 316/yr · 2.3yr | |
| Opencode (SST) | Agent IDEs & coding harnesses | Growth | 1.00 | 0.672 | 37111k | saturation expected ~Q2 2026 | 36 | 11,597,066 downloads · 36mo | |
| Alibaba Qwen 3 family | Foundation models (substrate reference) | Saturation | 1.00 | 0.131 | 5 | inflection passed Q2 2024 | 5 | 5 dated milestones | |
| Self-RAG | Retrieval-as-memory hybrids | Saturation | 1.00 | 0.114 | 18 | inflection passed Q2 2024 | 9 | 17 cites (5 infl) · 3yr | |
| Agent Workflow Memory | Recent method papers — theorized, no distinct product | Saturation | 1.00 | 0.341 | 10 | inflection passed Q1 2025 | 9 | 10 cites (2 infl) · 3yr | |
| HyDE | Retrieval-as-memory hybrids | Saturation | 1.00 | 0.144 | 12 | inflection passed Q2 2024 | 9 | 11 cites (0 infl) · 3yr | |
| LongMem | Recent method papers — theorized, no distinct product | Saturation | 1.00 | 0.168 | 12 | inflection passed Q3 2024 | 13 | 12 cites (0 infl) · 4yr | |
| FLARE | Retrieval-as-memory hybrids | Saturation | 1.00 | 0.152 | 11 | inflection passed Q4 2023 | 9 | 10 cites (0 infl) · 3yr | |
| PEFT (Hugging Face) | Training infrastructure | Saturation | 0.99 | 0.057 | 2.3k | inflection passed Q4 2024 | 31 | 1,691 commits · 31mo | |
| LLaMA-Factory | Training infrastructure | Saturation | 0.99 | 0.119 | 3.1k | inflection passed Q1 2024 | 21 | 3,023 commits · 21mo | |
| Memorizing Transformers | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.168 | 18 | inflection passed Q4 2023 | 9 | 17 cites (3 infl) · 3yr | |
| ExpeL | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.305 | 11 | inflection passed Q4 2024 | 9 | 11 cites (4 infl) · 3yr | |
| MemGPT v2 / agent-tools | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.276 | 30 | inflection passed Q1 2025 | 9 | 31 cites (10 infl) · 3yr | |
| S³ | Recent method papers — theorized, no distinct product | Growth | 0.99 | 0.076 | 34 | saturation expected ~Q2 2027 | 9 | 5 cites (1 infl) · 3yr | |
| MemoryBank | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.263 | 32 | inflection passed Q4 2024 | 13 | 32 cites (5 infl) · 4yr | |
| YaRN | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.229 | 12 | inflection passed Q4 2023 | 9 | 12 cites (3 infl) · 3yr | |
| Compressive Transformer | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.131 | 25 | inflection passed Q2 2023 | 21 | 24 cites (10 infl) · 6yr | |
| LongLoRA | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.276 | 8 | inflection passed Q1 2024 | 9 | 8 cites (0 infl) · 3yr | |
| xAI Grok 4 | Foundation models (substrate reference) | Saturation | 0.99 | 0.091 | 8 | inflection passed Q1 2025 | 7 | 7 dated milestones | |
| Cipher (byterover) (now ByteRover CLI) | Dedicated memory layers | Growth | 0.99 | 0.390 | 237k | saturation expected ~Q2 2026 | 36 | 71,148 downloads · 36mo | |
| TruLens | Evaluation & observability platforms | Saturation | 0.99 | 0.099 | 2.0k | inflection passed Q3 2024 | 27 | 1,736 commits · 27mo | |
| Anthropic Claude (foundation models) | Foundation models (substrate reference) | Saturation | 0.99 | 0.111 | 11 | inflection passed Q3 2025 | 9 | 9 dated milestones | |
| Episodic Memory Is the Missing Piece | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.273 | 46 | inflection passed Q1 2026 | 6 | 35 cites · 35/yr · 1.3yr | |
| LM2 | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.273 | 10 | inflection passed Q1 2026 | 6 | 8 cites · 8/yr · 1.3yr | |
| M+ (MemoryLLM) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.273 | 34 | inflection passed Q1 2026 | 6 | 26 cites · 26/yr · 1.3yr | |
| R³Mem | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.273 | 9 | inflection passed Q1 2026 | 6 | 7 cites · 7/yr · 1.3yr | |
| SoftCoT | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.273 | 42 | inflection passed Q1 2026 | 6 | 32 cites · 32/yr · 1.3yr | |
| Claude Code (Anthropic) | Agent IDEs & coding harnesses | Saturation | 0.99 | 0.385 | 402987k | inflection passed Q1 2026 | 36 | 287,847,897 downloads · 36mo | |
| Memobase | Dedicated memory layers | Saturation | 0.99 | 0.507 | 467 | inflection passed Q1 2025 | 14 | 467 commits · 14mo | |
| Transformer-XL | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.079 | 29 | inflection passed Q1 2024 | 29 | 25 cites (8 infl) · 8yr | |
| OpenAI Codex CLI | Agent IDEs & coding harnesses | Growth | 0.99 | 0.806 | 216905k | saturation expected ~Q2 2026 | 36 | 60,251,453 downloads · 36mo | |
| ReadAgent | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.236 | 8 | inflection passed Q1 2025 | 9 | 8 cites (1 infl) · 3yr | |
| AutoGen Memory | Framework-embedded memory | Saturation | 0.99 | 0.390 | 3.8k | inflection passed Q3 2024 | 14 | 3,782 commits · 14mo | |
| Agent S | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.390 | 353 | inflection passed Q1 2025 | 17 | 353 commits · 17mo | |
| 01.AI Yi family | Foundation models (substrate reference) | Saturation | 0.99 | 0.129 | 7 | inflection passed Q3 2024 | 5 | 5 dated milestones | |
| GPT Engineer | Agent frameworks (no first-party memory layer) | Saturation | 0.99 | 0.368 | 1.7k | inflection passed Q4 2023 | 13 | 1,648 commits · 13mo | |
| COLA | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.315 | 4 | inflection passed Q4 2025 | 5 | 3 cites · 3/yr · 1.2yr | |
| HiRAG | Retrieval-as-memory hybrids | Saturation | 0.99 | 0.315 | 32 | inflection passed Q4 2025 | 5 | 27 cites · 27/yr · 1.2yr | |
| MemInsight | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.315 | 58 | inflection passed Q4 2025 | 5 | 48 cites · 48/yr · 1.2yr | |
| RMM (Reflective Memory Management) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.315 | 60 | inflection passed Q4 2025 | 5 | 50 cites · 50/yr · 1.2yr | |
| OpenAI Realtime API | Voice agent platforms | Saturation | 0.99 | 0.873 | 44 | inflection passed Q2 2025 | 8 | 46 commits · 8mo | |
| RAPTOR | Retrieval-as-memory hybrids | Saturation | 0.99 | 0.226 | 15 | inflection passed Q4 2024 | 9 | 15 cites (5 infl) · 3yr | |
| REALM | Retrieval-as-memory hybrids | Saturation | 0.99 | 0.106 | 147 | inflection passed Q4 2020 | 44 | 147 commits · 44mo | |
| torchtune (PyTorch) | Training infrastructure | Saturation | 0.99 | 0.223 | 1.4k | inflection passed Q3 2024 | 22 | 1,339 commits · 22mo | |
| EMU (Efficient Episodic Memory Utilization) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.171 | 16 | inflection passed Q3 2025 | 9 | 14 cites · 7/yr · 2.2yr | |
| R2I | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.171 | 48 | inflection passed Q3 2025 | 9 | 42 cites · 21/yr · 2.2yr | |
| Distilabel | Training infrastructure | Saturation | 0.99 | 0.339 | 846 | inflection passed Q1 2024 | 20 | 846 commits · 20mo | |
| Neo4j | Knowledge-graph platforms | Saturation | 0.99 | 0.566 | 336 | inflection passed Q2 2025 | 14 | 345 commits · 14mo | |
| Differentiable Search Index (DSI) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.084 | 459 | inflection passed Q3 2024 | 18 | 405 cites · 101/yr · 4.3yr | |
| Streaming-LLM | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.223 | 19 | inflection passed Q1 2024 | 9 | 19 cites (0 infl) · 3yr | |
| Graphiti MCP Server (Zep) | Claude Code memory mechanisms | Saturation | 0.99 | 0.210 | 940 | inflection passed Q2 2025 | 22 | 829 commits · 22mo | |
| Zep & Graphiti | Dedicated memory layers | Saturation | 0.99 | 0.210 | 940 | inflection passed Q2 2025 | 22 | 829 commits · 22mo | |
| EWC (Elastic Weight Consolidation) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.062 | 10 | inflection passed Q4 2023 | 21 | 9 cites (3 infl) · 6yr | |
| Text Generation Inference (TGI) | Inference platforms & gateways | Saturation | 0.99 | 0.171 | 1.5k | inflection passed Q2 2024 | 28 | 1,450 commits · 28mo | |
| CREATOR | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.129 | 74 | inflection passed Q4 2024 | 13 | 70 cites · 24/yr · 3.0yr | |
| JARVIS-1 | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.144 | 203 | inflection passed Q2 2025 | 11 | 179 cites · 72/yr · 2.5yr | |
| ARMT (Associative RMT) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.107 | 487 | inflection passed Q3 2022 | 40 | 443 commits · 40mo | |
| RecurrentGPT | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.129 | 95 | inflection passed Q4 2024 | 13 | 90 cites · 30/yr · 3.0yr | |
| Dynamic Cheatsheet | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.339 | 65 | inflection passed Q4 2025 | 5 | 56 cites · 56/yr · 1.1yr | |
| From Human Memory to AI Memory (survey) | Theoretical / informal — ideas without a paper | Saturation | 0.99 | 0.339 | 85 | inflection passed Q4 2025 | 5 | 73 cites · 73/yr · 1.1yr | |
| GraphRAFT | Retrieval-as-memory hybrids | Saturation | 0.99 | 0.339 | 7 | inflection passed Q4 2025 | 5 | 6 cites · 6/yr · 1.1yr | |
| Inducing Programmatic Skills | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.339 | 49 | inflection passed Q4 2025 | 5 | 42 cites · 42/yr · 1.1yr | |
| LearnAct | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.339 | 33 | inflection passed Q4 2025 | 5 | 28 cites · 28/yr · 1.1yr | |
| SkillWeaver | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.339 | 63 | inflection passed Q4 2025 | 5 | 54 cites · 54/yr · 1.1yr | |
| UFO2 | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.339 | 34 | inflection passed Q4 2025 | 5 | 29 cites · 29/yr · 1.1yr | |
| Infini-attention | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.184 | 205 | inflection passed Q2 2025 | 9 | 192 cites · 96/yr · 2.1yr | |
| TransformerFAM | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.184 | 19 | inflection passed Q2 2025 | 9 | 18 cites · 9/yr · 2.1yr | |
| Adaptive-RAG | Retrieval-as-memory hybrids | Saturation | 0.99 | 0.184 | 479 | inflection passed Q2 2025 | 9 | 449 cites · 224/yr · 2.1yr | |
| SnapKV | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.171 | 71 | inflection passed Q2 2025 | 9 | 63 cites · 31/yr · 2.1yr | |
| Hugging Face LeRobot | Robotics foundation models & agent stacks | Growth | 0.99 | 0.049 | 11k | saturation expected ~Q3 2029 | 26 | 1,454 commits · 26mo | |
| MoT (Memory-of-Thought) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.137 | 58 | inflection passed Q4 2024 | 12 | 55 cites · 18/yr · 2.9yr | |
| Knowledge Graph Tuning | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.184 | 7 | inflection passed Q3 2025 | 9 | 6 cites · 3/yr · 2.0yr | |
| RAPTOR | Retrieval-as-memory hybrids | Saturation | 0.99 | 0.129 | 633 | inflection passed Q4 2025 | 10 | 452 cites · 226/yr · 2.3yr | |
| Genie | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.184 | 579 | inflection passed Q3 2025 | 9 | 511 cites · 255/yr · 2.0yr | |
| RET-LLM | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.114 | 68 | inflection passed Q1 2025 | 13 | 58 cites · 19/yr · 3.0yr | |
| I-JEPA | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.137 | 812 | inflection passed Q4 2024 | 12 | 773 cites · 257/yr · 2.9yr | |
| Memformer | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.160 | 75 | inflection passed Q1 2024 | 14 | 77 cites · 12/yr · 3.4yr | |
| Meta Llama 4 family | Foundation models (substrate reference) | Saturation | 0.99 | 0.075 | 8 | inflection passed Q4 2024 | 7 | 7 dated milestones | |
| Eliza (ai16z) | Agent frameworks (no first-party memory layer) | Saturation | 0.99 | 0.271 | 1091k | inflection passed Q4 2025 | 36 | 1,022,703 downloads · 36mo | |
| Toolformer | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.099 | 34 | inflection passed Q3 2024 | 13 | 30 cites (2 infl) · 4yr | |
| Mistral Large 2 / Mixtral family | Foundation models (substrate reference) | Saturation | 0.99 | 0.158 | 8 | inflection passed Q1 2024 | 8 | 8 dated milestones | |
| LoCoMo | Memory benchmarks & evaluation | Saturation | 0.99 | 0.223 | 38 | inflection passed Q2 2025 | 9 | 38 cites (9 infl) · 3yr | |
| LongRAG | Retrieval-as-memory hybrids | Saturation | 0.99 | 0.210 | 137 | inflection passed Q2 2025 | 8 | 128 cites · 64/yr · 1.9yr | |
| MATTER | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.210 | 4 | inflection passed Q2 2025 | 8 | 4 cites · 2/yr · 1.9yr | |
| MemoChat | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.158 | 73 | inflection passed Q4 2024 | 12 | 73 cites · 24/yr · 2.8yr | |
| EMAT | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.107 | 62 | inflection passed Q3 2024 | 15 | 58 cites · 14/yr · 3.6yr | |
| AndroidWorld | Memory benchmarks & evaluation | Saturation | 0.99 | 0.271 | 274 | inflection passed Q4 2024 | 21 | 274 commits · 21mo | |
| AiT (Associative Transformer) | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.144 | 3 | inflection passed Q1 2025 | 11 | 3 cites · 1/yr · 2.7yr | |
| Larimar | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.210 | 39 | inflection passed Q2 2025 | 8 | 37 cites · 18/yr · 1.9yr | |
| Lyfe Agents | Recent method papers — theorized, no distinct product | Saturation | 0.99 | 0.171 | 59 | inflection passed Q4 2024 | 11 | 59 cites · 19/yr · 2.6yr | |
| LoRA | Training infrastructure | Saturation | 0.98 | 0.076 | 17 | inflection passed Q2 2024 | 13 | 15 cites (6 infl) · 4yr | |
| Memoria (cognitive) | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.171 | 6 | inflection passed Q4 2024 | 11 | 6 cites · 2/yr · 2.6yr | |
| MemGuide | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.362 | 1 | inflection passed Q4 2025 | 5 | 1 cites · 1/yr · 1.0yr | |
| Alita | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.362 | 107 | inflection passed Q4 2025 | 5 | 92 cites · 92/yr · 1.0yr | |
| ATLAS | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.362 | 49 | inflection passed Q4 2025 | 5 | 42 cites · 42/yr · 1.0yr | |
| Darwin Gödel Machine | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.362 | 96 | inflection passed Q4 2025 | 5 | 82 cites · 82/yr · 1.0yr | |
| TC-RAG | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.223 | 30 | inflection passed Q2 2025 | 8 | 28 cites · 14/yr · 1.8yr | |
| WeKnow-RAG | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.223 | 15 | inflection passed Q2 2025 | 8 | 14 cites · 7/yr · 1.8yr | |
| Memory³ | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.210 | 42 | inflection passed Q2 2025 | 8 | 39 cites · 19/yr · 1.8yr | |
| OpenRLHF | Training infrastructure | Saturation | 0.98 | 0.144 | 1.6k | inflection passed Q3 2024 | 26 | 1,537 commits · 26mo | |
| Speculative RAG | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.236 | 97 | inflection passed Q2 2025 | 8 | 97 cites · 48/yr · 1.8yr | |
| CL of LLMs Survey | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.223 | 47 | inflection passed Q1 2024 | 13 | 47 commits · 13mo | |
| NVIDIA ReMEmbR | Vertical / domain-specific AI memory | Saturation | 0.98 | 0.223 | 67 | inflection passed Q2 2025 | 8 | 63 cites · 31/yr · 1.8yr | |
| PWM (Policy Learning with Large World Models) | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.236 | 11 | inflection passed Q2 2025 | 8 | 11 cites · 5/yr · 1.8yr | |
| Buffer of Thoughts | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.236 | 7 | inflection passed Q2 2025 | 8 | 7 cites · 3/yr · 1.9yr | |
| Microsoft Phi-4 family | Foundation models (substrate reference) | Saturation | 0.98 | 0.137 | 6 | inflection passed Q2 2024 | 6 | 6 dated milestones | |
| NetHack Learning Environment (NLE) | Memory benchmarks & evaluation | Saturation | 0.98 | 0.066 | 9.8k | inflection passed Q4 2016 | 52 | 9,823 commits · 52mo | |
| V-JEPA (original) | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.122 | 414 | inflection passed Q4 2025 | 10 | 259 cites · 129/yr · 2.3yr | |
| Fincon | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.672 | 118 | inflection passed Q3 2025 | 5 | 124 cites · 62/yr · 1.0yr | |
| DeepSpeed | Training infrastructure | Growth | 0.98 | 0.018 | 11k | saturation expected ~Q2 2030 | 28 | 3,142 commits · 28mo | |
| AGiXT Adaptive Memory | Framework-embedded memory | Growth | 0.98 | 0.031 | 12k | saturation expected ~Q2 2026 | 13 | 5,786 commits · 13mo | |
| KAG | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.739 | 1.3k | inflection passed Q1 2025 | 9 | 1,261 commits · 9mo | |
| LanceDB | Vector-database infrastructure | Growth | 0.98 | 0.040 | 6.7k | saturation expected ~Q2 2027 | 17 | 2,483 commits · 17mo | |
| RSL-RL | Training infrastructure | Growth | 0.98 | 0.091 | 362 | saturation expected ~Q4 2026 | 20 | 134 commits · 20mo | |
| Planning from Imagination | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.385 | 7 | inflection passed Q2 2025 | 6 | 7 cites · 3/yr · 1.4yr | |
| ChunkRAG | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.294 | 35 | inflection passed Q2 2025 | 7 | 35 cites · 17/yr · 1.6yr | |
| LLM-ACTR / Cognitive LLMs | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.317 | 8 | inflection passed Q3 2025 | 7 | 8 cites · 4/yr · 1.5yr | |
| Multi-Head RAG (MHRAG) | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.099 | 20 | inflection passed Q1 2026 | 6 | 10 commits · 6mo | |
| ToolGen | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.294 | 51 | inflection passed Q2 2025 | 7 | 51 cites · 25/yr · 1.6yr | |
| Bolt.new (StackBlitz) | Coding-agent memory | Saturation | 0.98 | 0.066 | 2479k | inflection passed Q4 2020 | 136 | 2,754,123 downloads · 136mo | |
| NAMMs | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.354 | 5 | inflection passed Q2 2025 | 7 | 5 cites · 2/yr · 1.6yr | |
| EM-LLM | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.294 | 52 | inflection passed Q2 2025 | 7 | 52 cites · 26/yr · 1.6yr | |
| Stagehand | Computer-use & desktop agents | Growth | 0.98 | 0.091 | 8.7k | saturation expected ~Q4 2027 | 16 | 1,286 commits · 16mo | |
| AI PERSONA | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.339 | 28 | inflection passed Q3 2025 | 6 | 28 cites · 14/yr · 1.4yr | |
| CAG (Cache-Augmented Generation) | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.339 | 52 | inflection passed Q3 2025 | 6 | 52 cites · 26/yr · 1.4yr | |
| Memory Layers at Scale | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.339 | 22 | inflection passed Q3 2025 | 6 | 22 cites · 11/yr · 1.4yr | |
| Titans (Google) | Research / specialised systems | Saturation | 0.98 | 0.385 | 200 | inflection passed Q3 2025 | 6 | 206 cites · 103/yr · 1.3yr | |
| Composer (MosaicML) | Training infrastructure | Saturation | 0.98 | 0.107 | 2.9k | inflection passed Q1 2023 | 28 | 2,701 commits · 28mo | |
| DeepSeek R1 / V3 family | Foundation models (substrate reference) | Saturation | 0.98 | 0.075 | 7 | inflection passed Q2 2025 | 6 | 6 dated milestones | |
| OpenAI Gym Retro | Memory benchmarks & evaluation | Saturation | 0.98 | 0.213 | 273 | inflection passed Q3 2018 | 28 | 281 commits · 28mo | |
| Voyager | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.137 | 17 | inflection passed Q4 2024 | 13 | 16 cites (4 infl) · 4yr | |
| CrewAI Memory | Framework-embedded memory | Growth | 0.98 | 0.044 | 23k | saturation expected ~Q2 2030 | 13 | 2,403 commits · 13mo | |
| RepairAgent | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.226 | 65 | inflection passed Q4 2024 | 12 | 62 commits · 12mo | |
| Recurrent Memory Transformer (RMT) | Recent method papers — theorized, no distinct product | Saturation | 0.98 | 0.114 | 406 | inflection passed Q2 2022 | 30 | 361 commits · 30mo | |
| StructRAG | Retrieval-as-memory hybrids | Saturation | 0.98 | 0.792 | 10 | inflection passed Q4 2024 | 5 | 10 commits · 5mo | |
| Cleanba | Training infrastructure | Saturation | 0.98 | 0.588 | 99 | inflection passed Q2 2023 | 10 | 101 commits · 10mo | |
| Physical Intelligence (π) | Robotics foundation models & agent stacks | Saturation | 0.98 | 0.381 | 224 | inflection passed Q3 2025 | 13 | 213 commits · 13mo | |
| Reka Core / Flash / Edge | Foundation models (substrate reference) | Growth | 0.98 | 0.044 | 48 | saturation expected ~Q3 2030 | 6 | 6 dated milestones | |
| ColossalAI | Training infrastructure | Saturation | 0.98 | 0.158 | 4.0k | inflection passed Q2 2023 | 23 | 3,977 commits · 23mo | |
| OASIS | Recent method papers — theorized, no distinct product | Saturation | 0.97 | 0.226 | 986 | inflection passed Q4 2024 | 25 | 939 commits · 25mo | |
| RLHFlow | Training infrastructure | Saturation | 0.97 | 0.351 | 151 | inflection passed Q3 2024 | 12 | 137 commits · 12mo | |
| DPO | Training infrastructure | Saturation | 0.97 | 0.168 | 10 | inflection passed Q2 2024 | 13 | 11 cites (1 infl) · 4yr | |
| RWKV-7 | Recent method papers — theorized, no distinct product | Saturation | 0.97 | 0.102 | 817 | inflection passed Q2 2023 | 56 | 875 commits · 56mo | |
| OpenLLMetry (Traceloop) | Evaluation & observability platforms | Saturation | 0.97 | 0.099 | 1.5k | inflection passed Q3 2024 | 27 | 1,385 commits · 27mo | |
| smolagents (Hugging Face) | Agent frameworks (no first-party memory layer) | Saturation | 0.97 | 0.452 | 1.0k | inflection passed Q1 2025 | 18 | 1,041 commits · 18mo | |
| smolagents Memory | Framework-embedded memory | Saturation | 0.97 | 0.452 | 1.0k | inflection passed Q1 2025 | 18 | 1,041 commits · 18mo | |
| ToolLLM | Recent method papers — theorized, no distinct product | Saturation | 0.97 | 1.01 | 176 | inflection passed Q3 2023 | 13 | 184 commits · 13mo | |
| Google Gemini 3 family | Foundation models (substrate reference) | Saturation | 0.97 | 0.099 | 7 | inflection passed Q1 2025 | 7 | 7 dated milestones | |
| Tantivy | Search platforms (non-memory) | Saturation | 0.97 | 0.038 | 3.7k | inflection passed Q2 2020 | 45 | 3,528 commits · 45mo | |
| GraphRAG (Microsoft) | Retrieval-as-memory hybrids | Saturation | 0.97 | 0.294 | 442 | inflection passed Q4 2024 | 21 | 465 commits · 21mo | |
| LazyGraphRAG | Retrieval-as-memory hybrids | Saturation | 0.97 | 0.294 | 442 | inflection passed Q4 2024 | 21 | 465 commits · 21mo | |
| Crafter | Memory benchmarks & evaluation | Saturation | 0.97 | 0.486 | 161 | inflection passed Q2 2021 | 15 | 161 commits · 15mo | |
| Burr (DAGWorks) | Agent frameworks (no first-party memory layer) | Saturation | 0.97 | 0.385 | 828 | inflection passed Q2 2024 | 29 | 920 commits · 29mo | |
| AlphaEdit | Recent method papers — theorized, no distinct product | Saturation | 0.97 | 0.605 | 39 | inflection passed Q1 2025 | 6 | 39 commits · 6mo | |
| LongMemEval | Memory benchmarks & evaluation | Saturation | 0.97 | 0.302 | 39 | inflection passed Q4 2024 | 9 | 40 commits · 9mo | |
| HELMET | Memory benchmarks & evaluation | Saturation | 0.97 | 0.407 | 95 | inflection passed Q4 2024 | 10 | 95 commits · 10mo | |
| Obsidian Smart Connections | Personal AI / PKM / lifelogging memory | Saturation | 0.97 | 0.122 | 2.1k | inflection passed Q1 2025 | 22 | 1,816 commits · 22mo | |
| pgvector | Vector-database infrastructure | Saturation | 0.97 | 0.190 | 1.8k | inflection passed Q4 2023 | 31 | 1,783 commits · 31mo | |
| Mem0 | Dedicated memory layers | Saturation | 0.97 | 0.129 | 2.3k | inflection passed Q4 2024 | 18 | 2,181 commits · 18mo | |
| Reflexion | Recent method papers — theorized, no distinct product | Saturation | 0.96 | 0.801 | 176 | inflection passed Q2 2023 | 10 | 176 commits · 10mo | |
| RETRO | Retrieval-as-memory hybrids | Saturation | 0.96 | 0.747 | 140 | inflection passed Q1 2022 | 8 | 140 commits · 8mo | |
| MemoRAG | Research / specialised systems | Saturation | 0.96 | 0.520 | 68 | inflection passed Q2 2024 | 5 | 68 commits · 5mo | |
| AutoRAG | Retrieval-as-memory hybrids | Saturation | 0.96 | 0.499 | 844 | inflection passed Q1 2024 | 25 | 881 commits · 25mo | |
| OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) | Foundation models (substrate reference) | Saturation | 0.96 | 0.053 | 10 | inflection passed Q2 2025 | 7 | 7 dated milestones | |
| MetaGPT | Recent method papers — theorized, no distinct product | Saturation | 0.96 | 0.470 | 6.4k | inflection passed Q2 2024 | 14 | 6,367 commits · 14mo | |
| MetaGPT | Agent IDEs & coding harnesses | Saturation | 0.96 | 0.470 | 6.4k | inflection passed Q2 2024 | 14 | 6,367 commits · 14mo | |
| Axolotl | Training infrastructure | Growth | 0.96 | 0.031 | 12k | saturation expected ~Q2 2029 | 18 | 2,744 commits · 18mo | |
| Stable-Baselines3 | Training infrastructure | Saturation | 0.96 | 0.057 | 933 | inflection passed Q1 2021 | 81 | 930 commits · 81mo | |
| NVIDIA GR00T / Isaac | Robotics foundation models & agent stacks | Saturation | 0.96 | 0.158 | 224 | inflection passed Q1 2026 | 15 | 132 commits · 15mo | |
| NVIDIA Isaac GR00T N1.7 | Vertical / domain-specific AI memory | Saturation | 0.96 | 0.158 | 224 | inflection passed Q1 2026 | 15 | 132 commits · 15mo | |
| Microsoft Semantic Kernel | Agent frameworks (no first-party memory layer) | Saturation | 0.96 | 0.122 | 5.3k | inflection passed Q1 2024 | 16 | 4,989 commits · 16mo | |
| Semantic Kernel Memory | Framework-embedded memory | Saturation | 0.96 | 0.122 | 5.3k | inflection passed Q1 2024 | 16 | 4,989 commits · 16mo | |
| A-MEM | Research / specialised systems | Saturation | 0.95 | 0.701 | 30 | inflection passed Q1 2025 | 8 | 32 commits · 8mo | |
| Memvid | Dedicated memory layers | Saturation | 0.95 | 2.99 | 258 | inflection passed Q4 2025 | 10 | 257 commits · 10mo | |
| Ragas | Evaluation & observability platforms | Growth | 0.95 | 0.053 | 11k | saturation expected ~Q2 2029 | 28 | 1,147 commits · 28mo | |
| Pokemon Red benchmark (speedrun / completion) | Memory benchmarks & evaluation | Saturation | 0.95 | 0.075 | 326 | inflection passed Q3 2023 | 33 | 272 commits · 33mo | |
| BGE-M3 | Retrieval-as-memory hybrids | Saturation | 0.95 | 0.294 | 1.5k | inflection passed Q3 2024 | 23 | 1,462 commits · 23mo | |
| Mem0 MCP (official) | Claude Code memory mechanisms | Saturation | 0.95 | 0.181 | 96 | inflection passed Q1 2026 | 7 | 53 commits · 7mo | |
| RULER | Memory benchmarks & evaluation | Saturation | 0.95 | 0.315 | 98 | inflection passed Q4 2024 | 13 | 101 commits · 13mo | |
| DreamerV3 | Recent method papers — theorized, no distinct product | Saturation | 0.95 | 0.053 | 47 | inflection passed Q4 2024 | 13 | 30 commits · 13mo | |
| BABILong | Memory benchmarks & evaluation | Saturation | 0.95 | 0.624 | 76 | inflection passed Q2 2024 | 9 | 84 commits · 9mo | |
| mcp-knowledge-graph | Claude Code memory mechanisms | Saturation | 0.94 | 0.226 | 37 | inflection passed Q1 2025 | 7 | 35 commits · 7mo | |
| TextWorld | Memory benchmarks & evaluation | Saturation | 0.94 | 0.099 | 593 | inflection passed Q3 2019 | 51 | 659 commits · 51mo | |
| AutoGen / AG2 | Agent frameworks (no first-party memory layer) | Growth | 0.94 | 0.013 | 16k | saturation expected ~Q4 2030 | 16 | 5,178 commits · 16mo | |
| Supermemory | Dedicated memory layers | Saturation | 0.93 | 0.049 | 2.8k | inflection passed Q1 2026 | 18 | 1,580 commits · 18mo | |
| Hanabi Learning Environment | Memory benchmarks & evaluation | Saturation | 0.93 | 0.368 | 28 | inflection passed Q2 2019 | 13 | 30 commits · 13mo | |
| Aider | Coding-agent memory | Saturation | 0.92 | 0.429 | 13k | inflection passed Q3 2024 | 13 | 13,133 commits · 13mo | |
| Neural Episodic Control | Recent method papers — theorized, no distinct product | Saturation | 0.92 | 0.591 | 38 | inflection passed Q2 2017 | 6 | 41 commits · 6mo | |
| Atlas | Retrieval-as-memory hybrids | Saturation | 0.91 | 0.181 | 23 | inflection passed Q2 2023 | 5 | 17 commits · 5mo | |
| LangMem (LangChain) | Framework-embedded memory | Saturation | 0.91 | 0.507 | 109 | inflection passed Q2 2025 | 13 | 121 commits · 13mo | |
| LongBench | Memory benchmarks & evaluation | Saturation | 0.90 | 0.520 | 67 | inflection passed Q3 2023 | 12 | 79 commits · 12mo | |
| HippoRAG / HippoRAG2 | Retrieval-as-memory hybrids | Growth | 0.84 | 0.076 | 620 | saturation expected ~Q1 2027 | 12 | 141 commits · 12mo | |
| Bonito | Training infrastructure | Decline | 0.78 | 0.249 | 37 | Q2 2024 | 6 | 35 commits · 6mo | |
| PROCGEN benchmark suite | Memory benchmarks & evaluation | Saturation | 0.76 | 0.195 | 75 | inflection passed Q1 2020 | 11 | 77 commits · 11mo | |
| π0.5 (Physical Intelligence) | Vertical / domain-specific AI memory | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| 11x.ai | Use-case-specific agent harnesses | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| 1X Technologies | Robotics foundation models & agent stacks | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| 6sense | Use-case-specific agent harnesses | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Abridge | Vertical / domain-specific AI memory | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| ACE — Agentic Context Engineering | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| ACON | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Acontext (memodb) | Dedicated memory layers | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Activeloop Deep Lake | Vector-database infrastructure | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Acuvity (now Proofpoint) | Memory governance, privacy & safety | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Adept ACT | Multi-agent orchestration platforms | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agency Swarm | Agent frameworks (no first-party memory layer) | Insufficient data | — | — | — | — | — | — | 2,412 commits · 9mo (under 12mo of history) |
| Agent KB | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agent QA | Use-case-specific agent harnesses | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AGENT-RECONFIGURE / Reconfigurable Agent | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agent Router (formerly Envoy AI Gateway) | Inference platforms & gateways | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agent2Agent Protocol (A2A) | Agent frameworks (no first-party memory layer) | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AgentBench | Memory benchmarks & evaluation | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AgentEvolver | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AgentFold | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| agentgateway | Inference platforms & gateways | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agentic Memory | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agentic Plan Caching (APC) | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| agentmemory (rohitg00) | Coding-agent memory | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AgentOps | Memory observability & monitoring | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AGENTS.md | File-backed / editor paradigms | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agility Robotics | Robotics foundation models & agent stacks | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Agno (Phidata) Memory | Framework-embedded memory | Insufficient data | — | — | — | — | — | — | 5,555 commits · 7mo (under 12mo of history) |
| AGNTCY (Internet of Agents) | Agent frameworks (no first-party memory layer) | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AI Singapore SEA-LION | Dedicated memory layers | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Aider CONVENTIONS.md / .aider.conf.yml | File-backed / editor paradigms | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Aider (harness) | Agent IDEs & coding harnesses | Insufficient data | — | — | — | — | — | — | 3,096,931 downloads · 6mo (under 12mo of history) |
| Aleph Alpha PhariaAI / PhariaCatch | Dedicated memory layers | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Algolia | Search platforms (non-memory) | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Alita-G | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AllegroGraph (Franz) | Knowledge-graph platforms | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| AlphaSense | Use-case-specific agent harnesses | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Amazon Bedrock Guardrails | Guardrails & AI security | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Amazon Neptune Analytics | Knowledge-graph platforms | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Amazon Nova family | Foundation models (substrate reference) | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Amazon Q Developer | Agent IDEs & coding harnesses | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Amplenote | Personal AI / PKM / lifelogging memory | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Anchor (Anchor Browser) | AI sandbox & runtime environments | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Anda Hippocampus (now Anda Brain) | Dedicated memory layers | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Anthropic Circuit Tracing | Recent method papers — theorized, no distinct product | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
| Anthropic Claude Memory | Platform-provider memory | Insufficient data | — | — | — | — | — | — | no parseable temporal signal |
Showing first 250 of 1002 rows. Narrow filters above or click a phase counter to focus.