Gemini 2.5 Pro (GA) · LOFT · Hard retrieval; average context length up to 128K; 300 queries · 成绩与配置

Gemini 2.5 Pro (GA) 的 LOFT · Hard retrieval; average context length up to 128K; 300 queries 记录。本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录),附 1 条来源定位。

成绩与评测条件

Gemini 2.5 Pro (GA) · LOFT 成绩与配置
指标与成绩配置与评测条件出处
Accuracy87%原文 87.0% · 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; 300 queries; average context length up to 128K.
条件
Gemini pass@1; single attempt unless stated otherwise.
在排名中查看此成绩

出处与披露

测试版本与指标

别名 LOFT (hard retrieval)

  • Accuracy · % · 越高越好