Skip to content
X Β· @teortaxesTex Β· X / Twitter

RT Maksym Andriushchenko @ ICML πŸ‡°πŸ‡·: i highly doubt that GLM-5.2 was benchmaxxed on PostTrainBench or heavily distilled from Claude models. anyon…

RT Maksym Andriushchenko @ ICML πŸ‡°πŸ‡·i highly doubt that GLM-5.2 was benchmaxxed on PostTrainBench or heavily distilled from Claude models. anyone can inspect the traces (https://posttrainbench.com/traces/):- the reasoning patterns overall look very reasonable. GLM-5.2 genuinely tries many very sensible approaches (see th