List of Flash News about massive bandwidth
| Time | Details |
|---|---|
| 19:07 |
OpenAI and Microsoft Co-Design AI Superfactory: Hundreds of Thousands of GPUs per Cluster and Massive Inter-Cluster Bandwidth for Scale
According to @gdb, OpenAI and Microsoft co-designed AI infrastructure featuring clusters scaled to hundreds of thousands of GPUs with massive bandwidth between clusters (Source: @gdb on X, Nov 15, 2025). He states that compute directly determines model capability and whether the team can scale to meet oversubscribed demand (Source: @gdb on X, Nov 15, 2025). He describes the system as an AI superfactory ready to be put to use (Source: @gdb on X, Nov 15, 2025). The post names Microsoft as co-designer, a material detail for equity traders monitoring MSFT’s AI infrastructure involvement, while providing no vendor, cost, or deployment timeline specifics (Source: @gdb on X, Nov 15, 2025). The post links to Satya Nadella’s thread but does not reference cryptocurrencies or blockchain, so no direct on-chain impact is specified (Source: @gdb on X, Nov 15, 2025). |