X · @teortaxesTex
· X / Twitter
more RL still needed Opus 4.6 is my favorite model btw. Not on economics, it's just the Pareto-optimal Opus. Low neuroticism, less sycophancy than 4.5…
more RL still neededOpus 4.6 is my favorite model btw. Not on economics, it's just the Pareto-optimal Opus. Low neuroticism, less sycophancy than 4.5 (which adored me disturbingly), smart, helpful. Brainwashed on Chyna but oh well.Bohu: 抽空做了个 Eval,评估下最近比较火的 GLM-5.2 能力先说结论: GLM-5.2 的模型能力(理解 + 写代码)不输 Opus 4.6,真正的差距在"如何在真