Gemini 2.5 Pro (GA) · Video-MMMU(Test subset) 83.6 · 附件证据(hist-t6-videommmu)

Gemini 2.5 Pro (GA) 的 Video-MMMU · Test subset 官方出处。Gemini 2.5 Pro stable API · 83.6。见《Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities — launch-day 70-page version》Historical PDF p.14, Table 6, VideoMMMU row, Gemini 2.5 Pro column。证据 hist-t6-videommmu。

原文与定位

Gemini 2.5 Pro (GA) · Video-MMMU(Test subset),Gemini 2.5 Pro stable API配置,Accuracy 83.6(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 — launch-day 70-page version》Historical PDF p.14, Table 6, VideoMMMU row, Gemini 2.5 Pro column。 查看成绩条件

Gemini 2.5 Pro / videommmu: 83.6. Test subset; 0-shot; max 256 frames.
官方材料
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities — launch-day 70-page version · PDF 第 14 页
定位
附件 · Historical PDF p.14, Table 6, VideoMMMU row, Gemini 2.5 Pro column
材料版本
70-page PDF, archived capture from 2025-06-17; cover dated 2025-06-17
版本公布
未确认
披露时间
· 已有首发依据
采集时间
历史依据
已实际下载首发当天官方PDF回放 https://web.archive.org/web/20250617164147id_/https://storage.googleapis.com/deepmind-media/gemini/gemini_v2_5_report.pdf;Memento-Datetime Tue, 17 Jun 2025 16:41:47 GMT。70/70页全文、安全章节及附录已核读,成绩表和图表已视觉核查;与现行73页版分开保存。
引用文本

Gemini 2.5 Pro (GA) · Video-MMMU(Test subset),Gemini 2.5 Pro stable API配置,Accuracy 83.6(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 — launch-day 70-page version(附件,70-page PDF, archived capture from 2025-06-17; cover dated 2025-06-17)Historical PDF p.14, Table 6, VideoMMMU row, Gemini 2.5 Pro column,https://web.archive.org/web/20250617164147id_/https://storage.googleapis.com/deepmind-media/gemini/gemini_v2_5_report.pdf#page=14。Benchlog https://benchlog.lekshome.cc/evidence/gemini-2-5-pro/hist-t6-videommmu/

关联成绩与条件

Gemini 2.5 Pro (GA) · Video-MMMU 成绩与配置
指标与成绩配置与评测条件证据关系
Accuracy83.6原文 83.6 · 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
Test subset; 0-shot; max 256 frames.
条件
Video sampled at 1 fps; video knowledge acquisition.

直接支持该成绩

在排名中查看此成绩