#201 - GPT 4.5, Sonnet 3.7, Grok 3, Phi 4
Last Week in AI
00:00
Advancements in Small Language Models
This chapter explores the latest advancements in Microsoft's small language models, focusing on the release of PHY4 and its multimodal capabilities. It features discussions on benchmarking tools like Sweelancer, the introduction of MAGMA, and issues related to specification gaming and model alignment. The conversation highlights the challenges and implications of AI model behavior, underscoring the importance of reliable evaluations and the evolving landscape of AI technology.
Transcript
Play full episode
