英伟达瞄准 AI 软件订阅收入,黄仁勋提出每块 GPU 每年约 2000 美元
黄仁勋近期在 BG2Pod 播客节目中表示,英伟达已进入医疗、汽车、通信和生命科学等行业,提供完整技术栈,并认为未来人们可能不再将其归类为芯片公司。
黄仁勋近期在 BG2Pod 播客节目中表示,英伟达已进入医疗、汽车、通信和生命科学等行业,提供完整技术栈,并认为未来人们可能不再将其归类为芯片公司。
Occurred:Unknown
黄仁勋近期在 BG2Pod 播客节目中表示,英伟达已进入医疗、汽车、通信和生命科学等行业,提供完整技术栈,并认为未来人们可能不再将其归类为芯片公司。
IT之家 科技新闻 ↗ Oct 26 09:41
By original publication time, with each report's bookmarks and feedback preserved.
Jensen Huang said licensing Nvidia's full software suite for deploying AI agents would cost about $2,000 per GPU per year. The company has not specified when the plan takes effect, which software it covers or which GPUs qualify. Nvidia also released the OpenShell reference architecture to limit agents' access to tools and resources; Anthropic combined it with Claude managed agents, while Salesforce developed a bridge for reviewing access requests in Slack.
Automatically verified against source evidence