Anthropic Reportedly Gray-Testing Upgraded Claude Opus 5.2 Model in Claude Code

Developers on social media report that Anthropic is gray-testing an unannounced "Claude Opus 5.2" model through backend routing in its CLI tool, Claude Code. Inspection of network request status logs revealed queries being routed to an `opus-5.2` model slug, delivering visibly faster response times and updated training data. If confirmed, this silent update indicates that Anthropic is significantly enhancing the speed, execution completeness, and agentic capabilities of its flagship model line. It highlights how top AI providers use silent canary deployments to test improved model architectures directly on developer workloads. Users discovered the model change after noticing that calling Opus via Claude Code produced substantially faster output and finished complex multi-step coding tasks without stopping to prompt for continuation. A community test prompt regarding recent niche events confirmed that the routed model possesses newer training weights compared to standard Opus 5.