Gemini 3 Pro · ScreenSpot-Pro(1,581 expert-annotated tasks across 23 professional applications and 3 operating systems) 60.5 · 附件证据(eval-screenspot-high)

Gemini 3 Pro 的 ScreenSpot-Pro · 1,581 expert-annotated tasks across 23 professional applications and 3 operating systems 官方出处。Gemini 3 Pro ScreenSpot-Pro media_resolution=high · 60.5。见《Gemini 3 Pro - Evaluations Approach, Methodology & Results》PDF p.1, Image methodology, ScreenSpot-Pro media_resolution setting note。证据 eval-screenspot-high。

原文与定位

Gemini 3 Pro · ScreenSpot-Pro(1,581 expert-annotated tasks across 23 professional applications and 3 operating systems),Gemini 3 Pro ScreenSpot-Pro media_resolution=high配置,原文数值(单位未说明) 60.5(具体成绩公布时间未确认)。见《Gemini 3 Pro - Evaluations Approach, Methodology & Results》PDF p.1, Image methodology, ScreenSpot-Pro media_resolution setting note。 查看成绩条件

With media_resoulution set to "high", Gemini 3.0 scores 60.5.
官方材料
Gemini 3 Pro - Evaluations Approach, Methodology & Results · PDF 第 1 页
定位
附件 · PDF p.1, Image methodology, ScreenSpot-Pro media_resolution setting note
材料版本
3-page evaluation PDF; results as of November 2025
版本公布
未确认
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
已实际取得首发当天官方PDF历史回放 https://web.archive.org/web/20251118163819id_/https://storage.googleapis.com/deepmind-media/gemini/gemini_3_pro_model_evaluation.pdf;回放原件与已完整核读的当前PDF完全相同。原件及响应记录:本轮整合目录 gemini-3-evaluation-historical-cdn.pdf。
引用文本

Gemini 3 Pro · ScreenSpot-Pro(1,581 expert-annotated tasks across 23 professional applications and 3 operating systems),Gemini 3 Pro ScreenSpot-Pro media_resolution=high配置,原文数值(单位未说明) 60.5(具体成绩公布时间未确认)。官方出处 Gemini 3 Pro - Evaluations Approach, Methodology & Results(附件,3-page evaluation PDF; results as of November 2025)PDF p.1, Image methodology, ScreenSpot-Pro media_resolution setting note,https://deepmind.google/models/evals-methodology/gemini-3-pro#page=1。Benchlog https://benchlog.lekshome.cc/evidence/gemini-3-pro/eval-screenspot-high/

关联成绩与条件

Gemini 3 Pro · ScreenSpot-Pro 成绩与配置
指标与成绩配置与评测条件证据关系
原文数值(单位未说明)60.5原文 60.5 · 单位 未标注
公布 未确认
Gemini 3 Pro ScreenSpot-Pro media_resolution=high
设置
Function calling with a capture-screenshot tool; media_resolution=high.
未知项
The method page says the high setting was coming soon to the API; no other settings are stated.
工具
允许工具
推理预算
未记录
框架
未记录
协议
Capture-screenshot function calling tool; media_resolution=high.
条件
The report does not print a percent sign for this alternate setting.

直接支持该成绩