🔢 Embedding System

Feature	Description	Performance
Multi-Provider	Agentic-Flow (ONNX), OpenAI, Transformers.js, Mock	4 providers
Auto-Install	ruflo embeddings init or createEmbeddingServiceAsync()	Zero config
75x Faster	Agentic-flow ONNX SIMD vs Transformers.js	3ms vs 230ms
Hyperbolic Space	Poincaré ball model for hierarchical data	Exponential capacity
Dimensions	384 to 3072 configurable	Quality vs speed tradeoff
Similarity Metrics	Cosine, Euclidean, Dot product, Hyperbolic distance	Task-specific matching
Neural Substrate	Drift detection, memory physics, swarm coordination	agentic-flow integration
LRU + SQLite Cache	Persistent cross-session caching	<1ms cache hits

bash
# Initialize ONNX embeddings with hyperbolic config
ruflo embeddings init

# Use larger model for higher quality
ruflo embeddings init --model all-mpnet-base-v2

# Semantic search
ruflo embeddings search -q "authentication patterns"

Mode	Adaptation	Quality	Memory	Use Case
real-time	<0.5ms	70%+	25MB	Production, low-latency
balanced	<18ms	75%+	50MB	General purpose
research	<100ms	95%+	100MB	Deep exploration
edge	<1ms	80%+	5MB	Resource-constrained
batch	<50ms	85%+	75MB	High-throughput

Algorithm	Type	Best For
PPO	Policy Gradient	Stable continuous learning
A2C	Actor-Critic	Balanced exploration/exploitation
DQN	Value-based	Discrete action spaces
Q-Learning	Tabular	Simple state spaces
SARSA	On-policy	Online learning
Decision Transformer	Sequence modeling	Long-horizon planning