Gemini 2.5 Pro (GA) · QVHighlights(Validation subset; [email protected]) 75 · 附件证据(t6-qvhighlights)

Gemini 2.5 Pro (GA) 的 QVHighlights · Validation subset; [email protected] 官方出处。Gemini 2.5 Pro stable API · 75。见《Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities》PDF p.15, Table 6, QVHighlights row, Gemini 2.5 Pro column。证据 t6-qvhighlights。

原文与定位

Gemini 2.5 Pro (GA) · QVHighlights(Validation subset; [email protected]),Gemini 2.5 Pro stable API配置,[email protected] 75(具体成绩公布时间未确认)。见《Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities》PDF p.15, Table 6, QVHighlights row, Gemini 2.5 Pro column。 查看成绩条件

QVHighlights 64.4 68.7 25.7 63.9 52.4 75.0 71.4
官方材料
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities · PDF 第 15 页
定位
附件 · PDF p.15, Table 6, QVHighlights row, Gemini 2.5 Pro column
材料版本
73-page PDF; embedded results include Gemini 2.5 Pro GA up to version 06-17; PDF metadata creation 2025-10-16 is not treated as publication date
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
暂无可靠历史版本依据
引用文本

Gemini 2.5 Pro (GA) · QVHighlights(Validation subset; [email protected]),Gemini 2.5 Pro stable API配置,[email protected] 75(具体成绩公布时间未确认)。官方出处 Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities(附件,73-page PDF; embedded results include Gemini 2.5 Pro GA up to version 06-17; PDF metadata creation 2025-10-16 is not treated as publication date)PDF p.15, Table 6, QVHighlights row, Gemini 2.5 Pro column,https://storage.googleapis.com/deepmind-media/gemini/gemini_v2_5_report.pdf#page=15。Benchlog https://benchlog.lekshome.cc/evidence/gemini-2-5-pro/t6-qvhighlights/

关联成绩与条件

Gemini 2.5 Pro (GA) · QVHighlights 成绩与配置
指标与成绩配置与评测条件证据关系
[email protected]75原文 75.0 · 单位 未标注
公布 未确认
Gemini 2.5 Pro stable API
设置
AI Studio API model id gemini-2.5-pro; default sampling settings. Technical report states Gemini scores are pass@1 and single-attempt unless specified. Aider Polyglot averages 3 trials; SWE-bench uses the internal agentic harness; Vibe-Eval uses Gemini as judge.
未知项
Thinking budget and any non-default settings for individual evaluations are not stated except where noted in each protocol.
工具
工具条件未记录
推理预算
未记录
框架
未记录
协议
Validation subset; 4-shot; [email protected]; max 256 frames.
条件
Video sampled at 1 fps.

直接支持该成绩

在排名中查看此成绩