WHAT NEW IN X10.29 THU

AI collections from X · 22 collections · 60,637 posts

minimax-m3.1POST

Parallel search-efficiency ranking puts MiniMax M3.1 Flash first at $43.8 per 1K tasks

Parallel's search-efficiency leaderboard ranks MiniMax M3.1 Flash first with a score of 68.8 at $43.8 per 1,000 tasks, ahead of Hy4 Preview, DeepSeek V4.1 Flash and GPT-6.1 Sol; the poster finds it good for research gathering but not proofreading.

鸡骑猫
@mspd4100Archived 8 days agoOriginally posted 2026.10.20
鸡骑猫 @mspd4100Originally posted 2026.10.20
Open original on X

看来 MiniMax M3.1 Flash 也不是完全没竞争力,做信息收集与整理应该还是适合的...审校和核对就算了 Parallel 的搜索效率排行榜将 MiniMax M3.1 Flash 列为第一,每千次任务成本仅 43.8 美元,得分 68.8,优于 Hy4 Preview、DeepSeek V4.1 Flash 和 GPT-6.1 Sol 等模型。 在智能分数与成本的散点图中,M3.1 Flash 以较低成本实现了较高性能,与 DeepSeek V4.1 Flash 的网页研究表现相当,但成本大约只有三分之一。

This post quotes another post: x.com/p0/status/2112544138662981825

x.com/mspd4100/status/2112571944449282193Archived 2026.10.21 12:05