Gemini 3.5 Flash-Lite · CharXiv Reasoning · Version/split not stated in selected source · 成绩与配置

Gemini 3.5 Flash-Lite 的 CharXiv Reasoning · Version/split not stated in selected source 记录。本次发布有 2 条成绩记录,含 2 条精确值(可含历史记录),附 2 条来源定位。

成绩与评测条件

Gemini 3.5 Flash-Lite · CharXiv Reasoning 成绩与配置
指标与成绩配置与评测条件出处
Accuracy74.5%原文 74.5% · 单位 %
公布 未确认
Gemini 3.5 Flash-Lite high thinking
设置
Gemini API gemini-3.5-flash-lite; high thinking; default sampling; pass@1 except where noted.
未知项
Exact thinking budget and sampling parameters are not published.
工具
禁止工具
推理预算
high
框架
未记录
协议
CharXiv Reasoning, no tools; self-computed.
条件
Gemini API, high thinking; split/version unstated.
在排名中查看此成绩
Accuracy76.5%原文 76.5% · 单位 %
公布 未确认
Gemini 3.5 Flash-Lite high thinking
设置
Gemini API gemini-3.5-flash-lite; high thinking; default sampling; pass@1 except where noted.
未知项
Exact thinking budget and sampling parameters are not published.
工具
允许工具
推理预算
high
框架
未记录
协议
CharXiv Reasoning with tools; Gemini uses search and code execution; self-computed.
条件
Gemini API, high thinking; split/version unstated.
在排名中查看此成绩

出处与披露

测试版本与指标

别名 CharXiv Reasoning

  • Accuracy · % · 越高越好