Overview
- Google released Gemini 4 Carbon and Anthropic announced Claude Fable 5.2 and gated Mythos 5.2 between Nov. 9 and Nov. 11, making the next wave of multimodal, long‑context models commercially available to developers and select verified users.
- Both vendors promoted big technical changes: Gemini 4 Carbon offers a 1,048,576‑token unified context window and API/Studio access, while Claude Fable 5.2 is available on major clouds with faster throughput, lower per‑token pricing and stronger anti‑distillation protections; Mythos 5.2 requires Cyber or Life Sciences verification for high‑risk use.
- Published benchmark results diverge across tests, with different models topping different leaderboards, and vendors highlighting cost‑efficiency or quality depending on the metric, leaving independent comparability and evaluation methods as active points of debate.
- Companies and investors are rethinking infrastructure plans because higher demand for inference, larger context windows and lower prices change where and how much capacity is needed, a shift that prompted SoftBank executives to say they may pause or move planned U.S. NeoCloud data‑center builds to other countries.
- Bloomberg reported OpenAI’s Q3 revenue jumped to about $10 billion, underscoring rapid commercial growth that is driving price cuts, broader cloud deployment and closer investor scrutiny of large, costly data‑center investments.