RAPTOR
https://github.com/parthsarthi03/raptor
Recursively embeds, clusters, and summarises chunks into a tree of abstractions. Retrieval traverses the tree at query time. ICLR 2024.
Research code · Verified 2026-10-23 · fresh as of 2026-10-23 · 28 cited facts, 1 with a third-party source · 1 recorded change · Suggest a correction
At a glance
- Type
- Recursive abstractive summary tree repo fix
- Tier
- T3
- Section
- Retrieval-as-memory hybrids
- Latest release
- no releases fix
- GitHub
- parthsarthi03/raptor — ~1.8k stars, 20 commits repo fix
- Pricing
- searched not found fix
- Funding
- not applicable — not commercial fix
- Freshness
- 2026-10-23 fix
- Layer
- agent-infra repo fix
- Licensing
- open-source repo fix
- Developer activity
- OpenRank 0.84 (12-mo avg 1.41) · 1 participants — 2026-01 third-party fix
Memory taxonomy
- storage
- vector
- retrieval
- similarity
- persistence
- long-term
- update
- read-only
- unit
- chunk
- governance
- inspectable
- conflict
- n/a
When to use
Optimised for: retrieval quality / accuracy on long-context QA
Anti-fit: most are research papers; production fitness varies by maintainer
Pros & cons
Pros
Hierarchical clustering retrieval — multi-resolution recall over long documents; widely adopted in 2024 RAG stacks.
Cons
Indexing cost grows with corpus; cluster quality depends on embedding choice.
Claims & capabilities
+20% absolute accuracy on QuALITY benchmark with GPT-4. Multi-step-reasoning state of the art on QA tasks at publication.
Technical surface
- API surface
- not applicable — research paper
- Backend storage
- not applicable — research paper
- Deployment
- searched not found
- Embedding model
- not applicable — research paper
- Multi-tenancy
- not applicable — research paper
- MCP
- not applicable — research paper, no deployed product
- A2A
- not applicable — research paper, no deployed product
- OpenTelemetry
- not applicable — research paper, no deployed product
Alternatives to RAPTOR → live options ranked by shared design, licence, hosting and MCP support
Similar systems
Other retrieval-as-memory hybrids in the catalog, ranked by inbound references.
- GraphRAG (Microsoft) T1
Modular graph-based RAG. Leiden community detection over knowledge graph; community summaries at multiple levels. DRIFT Search hybrid (Oct 2024). The GraphRAG repo is now largely in maintenance mode: no new features or PRs, only bug fixes and dependency/CVE updates.
- RETRO T3
DeepMind. Retrieval-Enhanced Transformer with chunked cross-attention over a 2T-token retrieval database. Frozen BERT retriever + differentiable encoder.
- Atlas T3
Meta AI. Jointly fine-tunes retriever + LM for few-shot tasks. The facebookresearch/atlas repo is no longer maintained; the code is provided as is.
- RAPTOR T3
Stanford recursive-summarisation tree over the corpus — clusters and summarises documents into a hierarchical index; retrieves at multiple levels. ICLR 2024.
- Self-RAG T4
Trains an LLM end-to-end to emit reflection tokens that decide whether to retrieve, assess passage relevance, and critique output quality — all within a single model.
- HippoRAG / HippoRAG2 T3
Hippocampus-inspired RAG. LLMs + knowledge graphs + Personalized PageRank to mimic neocortex/hippocampus roles. NeurIPS 2024.
Sources (27 cited facts, 1 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is Recursive abstractive summary tree repo
- What it does & what's distinct Recursively embeds, clusters, and summarises chunks into a tree of abstractions. Retriev… repo
- Claims / benefits +20% absolute accuracy on QuALITY benchmark with GPT-4. Multi-step-reasoning state of th… repo
- Created 2024-02 repo
- License MIT repo
- GitHub parthsarthi03/raptor — ~1.8k stars, 20 commits repo
- Citations 452 cites 226/yr paper
- Founders / pedigree Stanford NLP; Parth Sarthi, Salman Abdullah, Aditi Tuli, Shubh Khanna et al. (incl. Anna… paper
- Reproducibility Independently reproduced paper
- Code/weights release Code public + weights closed repo
- Versioning none repo
- Schema evolution schema with migration repo
- Validated verticals research, RAG developers (horizontal); some have production deployments repo
- Anti-fit most are research papers; production fitness varies by maintainer repo
- Optimised for retrieval quality / accuracy on long-context QA repo
- Adjacent infrastructure BYO LLM + vector store; OSS reference implementations exist repo
- Pros Hierarchical clustering retrieval — multi-resolution recall over long documents; widely … repo
- Cons Indexing cost grows with corpus; cluster quality depends on embedding choice. repo
- Project & sources GitHub parthsarthi03/raptor arxiv 2401.18059 OpenReview (ICLR) repo
- commit-trajectory [{"ym":"2024-02","cum":7},{"ym":"2024-03","cum":18},{"ym":"2024-06","cum":19},{"ym":"202… repo
- citation-trajectory [] paper
- architecture-deployment-locus local repo
- licensing-bucket open-source repo
- positioning-platform-vs-product platform repo
- taxonomy-three-layer-rollup agent-infra repo
- developer-activity-opendigger {"activity":2.02,"as_of":"2026-01","openrank_12mo":1.41,"openrank_current":0.84,"partici… third-party
- sdk-languages python repo
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-22
- Lifecycle first recorded as “research-complete”
Related systems
References (1)
- RETRO cites — S2 isInfluential citation