Gemini 3.7 Flash · Agents' Last Exam(Version not stated in selected source) 26.3% · 附件证据(ev-g37-ale)

Gemini 3.7 Flash 的 Agents' Last Exam · Version not stated in selected source 官方出处。Gemini 3.7 Flash default sampling · 26.3%。见《Gemini 3.7 Flash evaluation methodology and results》PDF p.4, Agent's Last Exam pass-rate row。证据 ev-g37-ale。

原文与定位

Gemini 3.7 Flash · Agents' Last Exam(Version not stated in selected source),Gemini 3.7 Flash default sampling配置,Pass rate 26.3%(具体成绩公布时间未确认)。见《Gemini 3.7 Flash evaluation methodology and results》PDF p.4, Agent's Last Exam pass-rate row。 查看成绩条件

Agent's Last Exam, pass rate | Gemini 3.7 Flash 26.3%.
官方材料
Gemini 3.7 Flash evaluation methodology and results · PDF 第 4 页
定位
附件 · PDF p.4, Agent's Last Exam pass-rate row
材料版本
Official 4-page PDF; method pp.2-3, results p.4; results as of August 2026; rendered and visually checked
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 3.7 Flash · Agents' Last Exam(Version not stated in selected source),Gemini 3.7 Flash default sampling配置,Pass rate 26.3%(具体成绩公布时间未确认)。官方出处 Gemini 3.7 Flash evaluation methodology and results(附件,Official 4-page PDF; method pp.2-3, results p.4; results as of August 2026; rendered and visually checked)PDF p.4, Agent's Last Exam pass-rate row,https://deepmind.google/models/evals-methodology/gemini-3-7-flash#page=4。Benchlog https://benchlog.lekshome.cc/evidence/gemini-3-7-flash/ev-g37-ale/

关联成绩与条件

Gemini 3.7 Flash · Agents' Last Exam 成绩与配置
指标与成绩配置与评测条件证据关系
Pass rate26.3%原文 26.3% · 单位 %
公布 未确认
Gemini 3.7 Flash default sampling
设置
Gemini API gemini-3.7-flash; default sampling; pass@1 except where noted.
未知项
Exact sampling parameters are not published.
工具
工具条件未记录
推理预算
未记录
框架
ALE-Claw
协议
Self-computed with default ALE-Claw harness; pass rate.
条件
Task version unstated; separate from v1.1 rubric-score results.

直接支持该成绩

在排名中查看此成绩