中国 AI 公司 DeepSeek 的 R1 模型不仅技术受关注,更引发了关于训练成本与推理效率的深层讨论。Jim Fan 指出这是"效率竞争"的开端,可能重塑全球 AI 投资格局。
@DrJimFan: "DeepSeek R1 has changed everything. Not just the model, but the economics. Training cost matters. Inference cost matters. China just proved they can compete on efficiency." — 15.6k likes
NVIDIA 收入结构历史性转折
NVIDIA 本季度推理收入首次超过训练收入,这是 AI 行业的分水岭。意味着 AI 重心从"如何训练更好的模型"转向"如何大规模部署和优化推理"。
"The 'vibe coding' era is here. People are building entire apps by just describing them to AI. This is the democratization of software development we dreamed of."
— Andrej Karpathy (@karpathy) 描述 AI 编程新时代 | 8.9k likes