Claude API: structured outputs support expands, and older models keep getting retired.
Production users: pin versions and stop ignoring deprecation notes.
What happened
Anthropic continues expanding structured output support while deprecating older models. The pattern is consistent: new capabilities arrive, old models get sunset with 3-6 month notice.
Why it matters
Structured outputs reduce parsing errors and make integration more reliable. Model deprecations allow Anthropic to focus resources on current generation. The pace reflects rapid AI development.
What changed
• Structured outputs support expanded to additional Claude variants
• Model deprecations continue (older models retired or scheduled to retire)
• Docs + console experience is being consolidated for developers
Who it affects
• API users
• Anyone shipping Claude in production
• RAG/agents teams
What to do now
• Pin model versions in production (avoid surprise changes)
• Track deprecation notices and set upgrade windows
• Prefer structured outputs for tool-calling and pipelines
Related updates
GPT-6 Sol and Luna broaden model choices
OpenAI introduced GPT-6 Sol and Luna as alternatives with different capability and efficiency balances.
GPT-Live-1 enters the voice API
OpenAI announced API access to GPT-Live-1 for full-duplex voice experiences.
Agent failures are mostly tool wiring, not model 'intelligence'.
Your agent isn't dumb. Your glue code is.
GPT-5.2 pushes harder on real work: code, tools, long context.
More useful for shipping work, especially with structure.
Claude Opus 4.5 leans into coding + agents, with stronger robustness.
Better for serious coding + agent work, especially when you wire it properly.
Prompt injection defenses are maturing (but still not bulletproof).
Defenses improved but aren't magic—implement defense-in-depth.