Samyama Graph is a Rust-native graph-vector database that lets developers store, query, search, and analyze connected data in one system.
It brings together graph traversal, OpenCypher-style querying, vector search, graph algorithms, and Redis-compatible access, making it useful for GraphRAG, knowledge graphs, AI agent memory, and large-scale relationship analytics.
Rust graph-vector database: OpenCypher (99.9% of evaluated TCK scenarios pass), vector search, graph algorithms, RESP + HTTP. LDBC SNB Interactive and BI complete; 1B edges on one machine.
Estrellas
187
Herramientas
—
Disponibilidad
—
Quickstart
Option 1 — Docker
Pulling the image needs no registry account: it is public on GitHub Container
Registry. latest is the newest release; pin a version such as :1.12.0 for a
reproducible setup.
Step 1 — Write docker-compose.yml in an empty folder:
services:
samyama-graph:
image: ghcr.io/samyama-ai/samyama-graph:latest
container_name: samyama-graph
restart: unless-stopped
ports:
- "6379:6379" # RESP (Redis protocol)
- "8080:8080" # HTTP API
environment:
# Lets the hosted visualizer (Step 4) call this server from your browser.
SAMYAMA_CORS_ORIGINS: https://graph.samyama.cloud
volumes:
- samyama-data:/app/samyama_data
volumes:
samyama-data:
Step 2 — Start it
docker compose up -d
docker logs -f samyama-graph # Ctrl-C to stop following the log
Runtime values go in params, never in the query text — they are bound, so a
value is never re-read as Cypher:
curl -X POST http://localhost:8080/api/query \
-H 'content-type: application/json' \
-d '{"query":"MATCH (p:Person) WHERE p.name = $name RETURN p.name","params":{"name":"Alice"}}'
Step 4 — Explore it visually (optional). Open the hosted visualizer at
https://graph.samyama.cloud/. The engine itself needs no account, but the
visualizer asks you to sign up or sign in. Then click Home, enter
http://localhost:8080 and click Connect. Without the SAMYAMA_CORS_ORIGINS
line from Step 1 the browser refuses the call — deliberately, because
/api/query runs arbitrary Cypher including DELETE
(#1328).
Step 5 — Stop or reset
docker compose down # stop
docker compose down -v # stop and delete all graph data in the volume
The server needs no credential by default. To require a bearer token, serve
TLS, or enable auto-embedding, see docs/DATA-HANDLING.md.
To load a real dataset, pick one from the case studies.
Option 2 — Build from source
zstd-sys generates its bindings with bindgen, which needs libclang; without it
the build fails part-way with a misleading 'stddef.h' file not found.
samyama-sdk = { git = "https://github.com/samyama-ai/samyama-graph", tag = "v1.12.0" } — not on crates.io yet
RemoteClient over HTTP, or EmbeddedClient::new() in-process
Any Redis client also works over RESP on port 6379. Details:
Python · TypeScript · Rust.
What can you build with Samyama Graph?
GraphRAG systems that combine vector search with graph traversal
Knowledge graph applications for enterprise, research, healthcare, and operations data
AI agent memory where entities, tools, actions, and context are stored as a graph
Biomedical and clinical graphs across papers, trials, pathways, drugs, and conditions
Fraud, investigation, infrastructure and dependency graphs for pattern and impact analysis
We loaded the entire PubMed corpus plus ClinicalTrials.gov, Reactome pathways, and DrugBank into one graph, then asked "What drugs are most tested in cancer clinical trials?"
MATCH (m:MeSHTerm)<-[:ANNOTATED_WITH]-(a:Article)
-[:REFERENCED_IN]->(t:ClinicalTrial)-[:TESTS]->(i:Intervention)
WHERE m.name = 'Neoplasms'
RETURN i.name, count(DISTINCT t) AS trials
ORDER BY trials DESC LIMIT 5
Drug
Trials
Placebo
521
Pembrolizumab
137
Carboplatin
106
Paclitaxel
106
Cyclophosphamide
98
10.3 seconds. One query. Four databases. 74 million nodes. 1 billion edges. A single machine.
That is query XK02 in
verified-results.csv:
10,250.2 ms, measured 2026-04-02 on one r6a.8xlarge and not re-measured since.
Provenance of each row is in docs/BENCHMARKS.md.
Cricket KG — 36K nodes, 1.4M edges, live graph simulation
Click for the full demo (1:56).
Case Studies — prove it yourself
case_studies/ downloads a real public knowledge graph, imports
it, runs showcase Cypher, and renders the session as a narrated GIF — one
command. Every showcase query is gated to return real rows
(Definition of Done).
A conformance harness publishes SCORECARD.json, and a regression gate blocks the release tag on a stale or red verdict
The 30-Second Tour
Cypher — MATCH, CREATE, MERGE, aggregations, path finding, 30+ functions.
99.9% of the openCypher TCK's evaluated scenarios pass (3,845 of 3,847, at 98.7%
coverage of the corpus, measured 2026-09-15) — see docs/CYPHER-COMPATIBILITY.md.
MATCH p = shortestPath((a:Person)-[:KNOWS*1..3]->(b:Person))
WHERE a.name = 'Alice'
RETURN b.name, length(p)
CALL pagerank('social') YIELD nodeId, score
RETURN nodeId, score ORDER BY score DESC LIMIT 10
Vector search — HNSW indexing for semantic search and GraphRAG; quantization: 'fp16' halves the index's memory.
CREATE VECTOR INDEX paper_idx FOR (p:Paper) ON (p.embedding) OPTIONS {dimensions: 384, similarity: 'cosine'}
CALL vector.search('Paper', 'embedding', [0.1, 0.2, 0.3], 10) YIELD node, score
Natural language — ask in English; an LLM translates to Cypher.
NLQ "Who are Alice's friends of friends that work at Google?"
AI agents — MCP servers generated from your graph schema: pip install samyama[mcp], then samyama-mcp-serve --demo cricket.
Out to pandas — any read query as Arrow or Parquet via POST /api/query/export (docs/BI-CONNECTIVITY.md).
Benchmarks
Run them with cargo bench --bench <name> (benches/); the vector,
optimization and micro/MVCC suites are self-contained, LDBC needs a data download.
Full results, provenance and caveats: docs/BENCHMARKS.md.
12/12 agree with the LDBC reference (CH-BENCH-GALX)
Cross-KG biomedical
see the table below
These run in-tree; LDBC certification is a formal third-party process we have not been through.
HIER (benchmarks/hier/) covers subsumption and hierarchical roll-up, which LDBC and FinBench do not. Latest: 108/108 agree — benchmarks/hier/results/PROVENANCE.json, engine commit 30d0731 with uncommitted changes ("dirty": true), committed 2026-09-21; 4 further corpus queries are specified but uncontrolled and are not in that denominator.
ID
Query
Time
First row (CSV)
XK02
Cancer → Trial interventions
10.3s
Placebo (521 trials)
XK07
Cancer trial sites by country
4.2s
United States (4,062)
XK08
NCI-funded → Trial interventions
20.5s
Placebo (933)
Cross-KG times from verified-results.csv, 2026-04-02, one r6a.8xlarge.
Examples and loaders
examples/ holds 127 programs: 125 Rust examples, a Python client
and an ESP32 sketch. Among them are 19 domain demos and 15 data loaders —
banking fraud, clinical trials, supply chain, manufacturing, SOC, LDBC,
FinBench, cricket, IMDB and more. Run the curated demos with
./scripts/run_all_examples.sh --batch, or one with cargo run --example banking_demo.
Guide:examples/README.md.
Everything in this repository is open source (Apache 2.0), including GPU
acceleration (wgpu, plus CUDA — build with --features gpu or --features cuda),
Prometheus metrics at /metrics with a Grafana dashboard, and a
request audit log (--audit-log). Samyama Enterprise adds:
Contributions are welcome — bug reports, docs, tests, and code. See
CONTRIBUTING.md for setup, build/test commands and the
pull request workflow. Found a bug? Open an issue.
Questions? Join the community chat.
License
Apache License 2.0 — use it in production, contribute back if you'd like.
Samyama (Sanskrit: संयम) — the union of focused query, sustained analysis, and unified insight.
S
Samyama Graph
samyama-ai·187 estrellas·
Rust graph-vector database: OpenCypher (99.9% of evaluated TCK scenarios pass), vector search, graph algorithms, RESP + HTTP. LDBC SNB Interactive and BI complete; 1B edges on one machine.
Aún no hay lista de herramientas. Los servidores remotos la reciben tras la primera comprobación de estado.
Preguntas frecuentes
Preguntas
¿Cómo instalo Samyama Graph?
Copia en esta página la configuración stdio para Cursor, Claude Desktop, Claude Code o VS Code. Usa la instalación del publicador, no un proxy de PlugMCP.
¿Samyama Graph necesita una clave de API?
Este registro no indica ninguna variable de entorno obligatoria.
¿Qué transporte usa Samyama Graph?
Samyama Graph se ejecuta localmente por stdio.
¿Cuándo se comprobó Samyama Graph por última vez?
Última comprobación: 2026-11-17. Rust graph-vector database: OpenCypher (99.9% of evaluated TCK scenarios pass), vector search, graph algorithms, RESP + HTTP. LDBC SNB Interactive and BI complete; 1B edges on one machine.