Model Registry/Claude 3.7 Sonnet

Claude 3.7 Sonnet

Approved

Anthropic · 2025-06 · Transformer LLM · Active · Assessed Sep 29, 2026

97

Overall Fit

97.4%

Accuracy

98.9%

Safety

97.8%

Reliability

98.5%

Compliance

510ms

Latency p95

Primary use case

Deep Reasoning & Compliance Analysis

Best suited for

Sensitive advisory workflows that need strong instruction following and safer responses.

LLMExtended ThinkingHigh SafetyNuance

Governance

Active guardrails
5
Decision status
Approved
Lifecycle
Active

Drift index, cost efficiency, and per-suite evidence are available inside the workspace.