Pause
Reset
GOVERNANCE (SAFETY/OPS)
APPLICATION (AGENTS/RAG)
RUNTIME (INFERENCE/SERVING)
FOUNDATION (TENSORS/MODELS)
Request
Generate
Context
Cache Hit
Decode
Similarity
POLICY ENGINE
AGENT ORCHESTRATOR
RAG PIPELINE
INFERENCE API
KV CACHE
VECTOR STORE
GPU HARDWARE
BASE MODELS
EMBEDDINGS
TRAINING DATA