3 results for “V7”
OpenAI published a case study on how V7's enterprise agent platform uses GPT-5.6 Luna, Terra and Sol plus GPT-6 Astra. V7 says the tiered setup cut tool-call errors to 0.2% and lowered per-document costs.
NVIDIA submitted its first MLPerf Inference v6.1 preview results for Vera Rubin NVL72, showing up to 3.7x higher throughput than GB300 NVL72 on Qwen3-VL. The numbers signal a leap for rack-scale inference on MoE models.
Sam Altman told TIME — and revisited the theme on the Sources podcast — that OpenAI will “definitely” make humanoid robots and that everyone should eventually have a personal robot. Separately, Bloomberg-reported details sketch a screenless, puck-sized Ive device aimed at 2027, while Apple’s trade-secret suit hangs over the hardware calendar.