I don’t understand why Murata would say this. Are they basically saying, “We’re not going to expand capacity, so we’ll just let others take our market share”?
We are committed to pushing the model frontier across cost efficiency, capability, and speed.
Starting today, we are reducing prices for GPT-5.6 Luna by 80% and GPT-5.6 Terra by 20% , and offering a faster option for GPT-5.6 Sol in the API.
Luna and Terra’s lower prices are reflected in how usage is counted in Codex and ChatGPT Work, so your usage goes further.
J.P摩根现在认为杠杆型ETF的平仓过程已经结束,对冲基金也已完成约90%的去杠杆化(两者均达到了可接受的水平)。
J.P.Morgan believes that leveraged ETF unwind is complete and hedge funds are done with ~90% of de-leveraging (both to acceptable levels).
J.P.Morgan 的报告表明:代币支出持续增长,GPU价格依旧坚挺,内存通胀开始降温。最要命的就是最后一句了。
7月份,按代币使用量排名前五的模型依次为:1) MiMo v2.5;2) DeepSeek v4 Flash;3) GLM 5.2;4) DeepSeek v4 Pro;5) MiniMax M3,它们合计占总代币使用量的50%。然而,若按支出金额计算,7月份排名前五的模型则为:1) Claude Opus 4.8;2) Claude Opus 4.7;3) Fable 5;4) Kimi K3;5) GPT-5.6 Sol,它们合计占总代币支出金额的54%。
两个排名显然还是支出金额的排名更能体现实力。