李飞飞团队以不到50美元的云计算成本, 使用16块H100显卡, 在26分钟内对阿里云的Qwen2.5-32B-Instruct模型进行监督微调, 成功训练出名为s1的AI推理模型. S1模型在数学和编码能力测试中表现出色, 媲美OpenAI的o1和DeepSeek的R1模型.
原理大概来说, 就是通过在模型生成过程中插入“等待”(Wait)步骤, 迫使模型在测试时充分利用额外的计算资源, 由此改善对复杂问题(如数学竞赛题目)的解题能力. 之前用别人模型叫蒸馏, 现在叫测试时扩展. 而且问题来了, 阿里云怎么会有H100的显卡. 以后AI采取高端寄居寄生技术, 人类的人工智能时代来临.
原理大概来说, 就是通过在模型生成过程中插入“等待”(Wait)步骤, 迫使模型在测试时充分利用额外的计算资源, 由此改善对复杂问题(如数学竞赛题目)的解题能力. 之前用别人模型叫蒸馏, 现在叫测试时扩展. 而且问题来了, 阿里云怎么会有H100的显卡. 以后AI采取高端寄居寄生技术, 人类的人工智能时代来临.
Reposted from
TechCrunch
AI researchers were able to train an AI “reasoning” model for under $50 in cloud compute credits, according to a new research paper.
The model, known as s1, performs similarly to cutting-edge reasoning models on tests measuring math and coding abilities.
Read more: tcrn.ch/4gvhxKA
The model, known as s1, performs similarly to cutting-edge reasoning models on tests measuring math and coding abilities.
Read more: tcrn.ch/4gvhxKA
Comments