News for hackers. Stuff that matters.
Re:Muse Spark: Scaling Towards Personal Superintelligence by nickvec on Thursday April 09, 2026 @05:08AM (#47701070)
The only benchmark I care about! Just curious Simon - which model do you think has created the best pelican riding a bicycle thus far?
Re:Muse Spark: Scaling Towards Personal Superintelligence by simonw on Thursday April 09, 2026 @04:46PM (#47709777)
Gemini 3.1 Pro: https://simonwillison.net/2026/Feb/19/gemini-31-pro/

But GLM-5.1 has the best NORTH VIRGINIA OPOSSUM ON AN E-SCOOTER: https://simonwillison.net/2026/Apr/7/glm-51/