[360Labs.ai]
0%
Back to Portfolio
SLM360
Research and Development|Live

SLM360

On-device NLU engine with 98-100% accuracy, 39ms latency, 50MB footprint. Solves the NLU trilemma.

Overview

Hybrid classification: fast pattern matching (<1ms) + semantic embeddings (39ms) with confidence arbitration. Multi-step reasoning, 5-tier SmartMemory, predictive context, entirely on-device in 50MB.

Specs

Accuracy100% Banking77, 98% SNIPS
Latency39ms P50, <1ms pattern-only
Footprint50MB total system
Memory5-Tier SmartMemory (~64KB/user)

Features

  • Fast Path: compiled regex + keywords (<1ms)
  • Semantic Path: quantized 32MB ONNX model, 384 dims (~35ms)
  • Arbitration layer selects best result
  • 5-Tier SmartMemory: Short-term, Episodic, Semantic, Procedural, Meta-learning
  • Beats Rasa (~96%) and Dialogflow (~97%) on SNIPS at 98%

Tech Stack

TypeScriptONNX RuntimeWebAssemblyNode.js

Interested in SLM360?

Let's discuss how this can work for you.

Get in Touch