Inkling, mentioned 4 times across podcast episodes and expert conversations analyzed by Sonic.
On the Humanity's Last Exam benchmark, the Inkling model scores 29.7% without tools and 46% with tools.
Thinking Machines Lab has released its first large language model, Inkling, which is a 975 billion parameter Mixture of Experts model.
Professor Ethan Mollick found the Inkling model to be "pretty rough" in his tests and not competitive with frontier Chinese open-weight models.