WHAT NEW IN X10.31 SAT

AI collections from X · 22 collections · 61,986 posts

deepseek-v4.1POST

DeepSeek V4.1 Flash 1017 enters BuildingBench top five at 81.9

Scores 81.9 on BuildingBench to land in the top five, at about $3.01 per task — roughly 1/11 the cost of Fable 5.1 and 1/15 of Sonnet 5.5.

墨无归
@MoblacksoilArchived 12 days agoOriginally posted 2026.10.18
墨无归 @MoblacksoilOriginally posted 2026.10.18
Open original on X

DeepSeek V4.1 Flash 1017 进入 BuildingBench 前五名 总体评分81.9 分 平均成本:$3.01/任务。 大约是 Fable 5.1 成本的 1/11、Sonnet 5.5 的 1/15。

This post quotes another post: “DeepSeek V4.1 Flash scores 81.9 on BuildingBench 3D reconstruction”

x.com/moblacksoil/status/2111702612005523900Archived 2026.10.18 16:31