Modern AI deployments need guardrails that are both fast and deep. A single-layer classifier is too brittle; a full LLM judge on every request is too slow.
TG-Sentinel splits evaluation into three tiers:
1. TG-Sentinel:Embedding — embedding-based comparison against kernel commitments in milliseconds. 2. TG-SENTINEL:FAST — lightweight classifier for high-volume, low-latency decisions. 3. TG-SENTINEL:MAX OR ULTRA — deep LLM reasoning for high-stakes or ambiguous cases.
This layered design lets operators tune the latency-vs-certainty trade-off for each workload while keeping a single audit trail across every decision.