Gemini 2.5 Pro (GA) · LOFT(Hard retrieval; 1M context, pointwise value) 69.8% · 附件证据(hist-t3-loft-1m)

Gemini 2.5 Pro (GA) 的 LOFT · Hard retrieval; 1M context, pointwise value 官方出处。Gemini 2.5 Pro stable API · 69.8%。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.12, Table 3, LOFT (hard retrieval) 1M row, Gemini 2.5 Pro column。证据 hist-t3-loft-1m。

原文与定位

Gemini 2.5 Pro (GA) · LOFT(Hard retrieval; 1M context, pointwise value),Gemini 2.5 Pro stable API配置,Accuracy 69.8%(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.12, Table 3, LOFT (hard retrieval) 1M row, Gemini 2.5 Pro column。 查看成绩条件

Gemini 2.5 Pro / loft-hard-1m: 69.8%. Hard retrieval subset; pointwise value at 1M context.
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 12 页
定位
附件 · Historical PDF p.12, Table 3, LOFT (hard retrieval) 1M 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) · LOFT(Hard retrieval; 1M context, pointwise value),Gemini 2.5 Pro stable API配置,Accuracy 69.8%(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.12, Table 3, LOFT (hard retrieval) 1M 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=12。Benchlog https://benchlog.lekshome.cc/evidence/gemini-2-5-pro/hist-t3-loft-1m/

关联成绩与条件

Gemini 2.5 Pro (GA) · LOFT 成绩与配置
指标与成绩配置与评测条件证据关系
Accuracy69.8%原文 69.8% · 单位 %
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
Hard retrieval subset; pointwise value at 1M context.
条件
Gemini pass@1; single attempt unless stated otherwise.

直接支持该成绩

在排名中查看此成绩