Gemini 3.1 Pro · ARC-AGI-2(ARC Prize Verified · semi-private set) 77.1% · 主文证据(main-arc-agi-2)

Gemini 3.1 Pro 的 ARC-AGI-2 · ARC Prize Verified · semi-private set 官方出处。Gemini 3.1 Pro Thinking (High) · 77.1%。见《Gemini 3.1 Pro: A smarter model for your most complex tasks》Release article / Building on Gemini 3 series / ARC-AGI-2 paragraph。证据 main-arc-agi-2。

原文与定位

Gemini 3.1 Pro · ARC-AGI-2(ARC Prize Verified · semi-private set),Gemini 3.1 Pro Thinking (High)配置,Accuracy 77.1%(具体成绩公布时间未确认)。见《Gemini 3.1 Pro: A smarter model for your most complex tasks》Release article / Building on Gemini 3 series / ARC-AGI-2 paragraph。 查看成绩条件

On ARC-AGI-2, a benchmark that evaluates a model's ability to solve entirely new logic patterns, 3.1 Pro achieved a verified score of 77.1%.
官方材料
Gemini 3.1 Pro: A smarter model for your most complex tasks
定位
主文 · Release article / Building on Gemini 3 series / ARC-AGI-2 paragraph
材料版本
Article page collected 2026-11-10; article date 2026-02-19
版本公布
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
暂无可靠历史版本依据
引用文本

Gemini 3.1 Pro · ARC-AGI-2(ARC Prize Verified · semi-private set),Gemini 3.1 Pro Thinking (High)配置,Accuracy 77.1%(具体成绩公布时间未确认)。官方出处 Gemini 3.1 Pro: A smarter model for your most complex tasks(主文,Article page collected 2026-11-10; article date 2026-02-19)Release article / Building on Gemini 3 series / ARC-AGI-2 paragraph,https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-1-pro/。Benchlog https://benchlog.lekshome.cc/evidence/gemini-3-1-pro/main-arc-agi-2/

关联成绩与条件

Gemini 3.1 Pro · ARC-AGI-2 成绩与配置
指标与成绩配置与评测条件证据关系
Accuracy77.1%原文 77.1% · 单位 %
公布 未确认
Gemini 3.1 Pro Thinking (High)
设置
Gemini API model id gemini-3.1-pro-preview; default sampling settings; evaluation table column specifies Thinking (High). All scores are pass@1 except where noted.
未知项
Exact thinking budget and sampling parameters are not listed.
工具
工具条件未记录
推理预算
high
框架
未记录
协议
ARC Prize Verified; semi-private set.
条件
Gemini 3.1 Pro Thinking (High); ARC Prize source.

直接支持该成绩

在排名中查看此成绩