Curated AI news from the last 24 hours — models, agents, enterprise, and developer tools.
TechCrunch reports Anthropic's Opus 5.5 matches Claude Fable 5.1 on most work while costing 40% less to run than Opus 5. Pricing is $4 input and $20 output per million tokens, and Sonnet 5.5 and Haiku 5.5 are due in the coming weeks.
techcrunch.com ↗DeepMind chief Koray Kavukcuoglu confirmed Gemini 4 has entered post-training, with safety testing under way and internal use powering Antigravity. He wants an early post-training release much sooner than the end of 2026, though no date was given.
9to5google.com ↗CNBC covers Anthropic and OpenAI releasing lower-priced models in their first launches since a call for a slowdown in frontier AI development. The releases signal competition shifting toward cost and efficiency.
cnbc.com ↗Microsoft launched a unified Copilot app combining chat, coding and long-running agents for business and personal use. Users can pick models from OpenAI or Anthropic, or let an Auto mode choose per request.
geekwire.com ↗