|
沙發

樓主 |
發表於 2025-8-24 16:48:36
|
只看該作者
The emergence of reasoning AI models has fundamentally changed the compute landscape. Microsoft processed over 100 trillion tokens in Q1, a fivefold YoY increase. DeepSeek and similar reasoning models generate 13 times more tokens than traditional AI models, creating exponential inference demand. NVIDIA's GB200 NVL72 delivers 30 times higher inference throughput, making it the ideal platform for this compute-intensive workload. |
|