netrunner(@plotarmordev):Meta's big model drop pulled the attention away from @StepFun_ai and everything else that shipped this week. But the thing is that a lot of frontier models look tuned to the benchmarks they get judged on, and a release chart can't tell you whether that's what you're looking at. I personally like to see the same tasks and the same harness, on the hardware people actually run.

netrunner

@plotarmordev

Cofounder & engineer. Running local AI, building things with it, and sharing what I learn.

参加 April 2022

806 フォロー中 8.4K ファン

netrunner@plotarmordev

2026.10.16 11:55

Meta's big model drop pulled the attention away from @StepFun_ai and everything else that shipped this week. But the thing is that a lot of frontier models look tuned to the benchmarks they get judged on, and a release chart can't tell you whether that's what you're looking at. I personally like to see the same tasks and the same harness, on the hardware people actually run.