API — Community free · Pro $29/mo

Semantic search that
actually explains itself.

Prime factorization algebra instead of black-box embeddings. 28× faster than cosine similarity. Deterministic, auditable, and LangChain-ready.

📖 API Docs (Swagger) View pricing
Φ(A) = 2 × 3 × 5
GCD(Φ(A), Φ(B)) = shared features
LCM(Φ(A), Φ(B)) = A ∪ B

The core idea

Every concept becomes
a prime number product

Instead of "King ≈ Queen (cosine 0.87)", we give you: "King = 2×3×5, Queen = 2×5×7. They share {2,5}."

# Install the SDK pip install neurosym-cloud # Use it from neurosym_cloud import TriadicClient client = TriadicClient(api_key="your_key") # Encode concepts to prime integers result = client.encode(["machine learning", "neural network", "backpropagation"]) # Search for similar concepts similar = client.search(result.primes[0], top_k=5) # Audit two embedding models discrepancies = client.audit(model_a="openai", model_b="cohere", concepts=[...])

Performance

28× faster than cosine

Verified across 50,000 pairwise operations. No approximations.

Triadic Engine
28.4× faster
Cosine similarity
baseline
Feature Triadic Engine Cosine similarity BM25
Deterministic results✗ (seed-dependent)
Explainable output✓ (shared primes)✗ (black box)
Composition algebra✓ (LCM)
Subsumption check✓ (GCD mod)
Model auditing✓ (108K discrepancies/2M)
Speed (50K ops)✓ 28× fasterbaseline~3× faster

API Reference

Core endpoints

🔢

POST /encode

Convert concepts → prime integers. 4 projection modes: random, PCA, consensus, contrastive.

🔍

POST /search

GCD similarity search. O(F×B) instead of O(N²). Returns top-k similar concepts with shared factors.

🔬

POST /audit

Compare two embedding models structurally. Found 108K discrepancies across 2M semantic chains.

⚠️

POST /anomaly/scan

Detect multiplicative rule violations in your data. Severity: CLEAN, INFO, WARNING, CRITICAL.

📁

Namespaces (Pro)

Persistent indices per user/project. Encode once, search forever. Up to 10 namespaces on Pro.

Batch jobs (Enterprise)

Async CSV processing for millions of concepts. Poll status with /batch/{job_id}.

Pricing

Start free,
scale when ready

Community
$0/mo

Perfect for exploring

  • 100 API calls / day
  • 2,000 concepts / request
  • /encode, /search, /audit
  • Swagger docs access
  • Community support
Start free
Enterprise
$299/mo

Annual contracts available

  • 1,000,000 API calls / day
  • Unlimited concepts
  • Unlimited namespaces
  • Batch processing (CSV)
  • Custom embedding models
  • On-premise license option
  • SLA + dedicated support
Contact sales

Integrations

Plug into your
existing AI stack

🦜

LangChain

Drop-in replacement for VectorStoreRetriever. Use Triadic as your RAG retrieval backend in any LangChain chain.

from neurosym_cloud.integrations import TriadicRetriever retriever = TriadicRetriever(client, namespace="my_docs")
🦙

LlamaIndex

Use Triadic as a custom retriever in LlamaIndex query engines. Compatible with the standard retriever interface.

from neurosym_cloud.integrations import TriadicLlamaRetriever retriever = TriadicLlamaRetriever(client)