AA-Briefcase · v1.1 · 测试定义与成绩

AA-Briefcase · v1.1 收录 6 次模型发布的评测披露或成绩。

测试版本与指标

别名 AA-Briefcase v1.1

  • Elo rating · Elo · 越高越好
  • Rubric score · % · 越高越好

各系列最新模型 · 公布成绩

常规展示范围。排名按指标与评测协议分组,已记录的工具、预算及框架分别展示。条件未说明的组保留公布值供查证。

Elo rating Elo

条件未说明

AA-Briefcase · v1.1 · Elo rating · 条件未说明
模型与配置成绩(Elo)记录与证据
Qwen4-Max未注明1686 Elo

AA-Briefcase v1.1 Elo rating。

AA-Briefcase · v1.1 · Elo rating · AA-Briefcase v1.1 Elo rating。
模型与配置成绩(Elo)记录与证据
Claude Fable 5.2未注明1785 Elo

工具条件未记录 / xhigh / 框架未记录 · AA-Briefcase v1.1; Elo rating.

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / xhigh / 框架未记录 · AA-Briefcase v1.1; Elo rating.
模型与配置成绩(Elo)记录与证据
Claude Fable 5.2Xhigh1738 Elo

工具条件未记录 / max / 框架未记录 · AA-Briefcase v1.1。

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / max / 框架未记录 · AA-Briefcase v1.1。
模型与配置成绩(Elo)记录与证据
Claude Haiku 5.5Max1578 Elo

工具条件未记录 / Medium / 框架未记录 · AA-Briefcase v1.1; long-horizon knowledge-work projects with linked tasks and source files.

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / Medium / 框架未记录 · AA-Briefcase v1.1; long-horizon knowledge-work projects with linked tasks and source files.
模型与配置成绩(Elo)记录与证据
Claude Haiku 5.5Medium1372 Elo

工具条件未记录 / max / 框架未记录 · AA-Briefcase v1.1 long-horizon knowledge-work benchmark; independent Artificial Analysis evaluation.

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / max / 框架未记录 · AA-Briefcase v1.1 long-horizon knowledge-work benchmark; independent Artificial Analysis evaluation.
模型与配置成绩(Elo)记录与证据
Claude Sonnet 5.5Max1811 Elo

工具条件未记录 / xhigh / 框架未记录 · AA-Briefcase v1.1 long-horizon knowledge-work benchmark; independent Artificial Analysis evaluation.

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / xhigh / 框架未记录 · AA-Briefcase v1.1 long-horizon knowledge-work benchmark; independent Artificial Analysis evaluation.
模型与配置成绩(Elo)记录与证据
Claude Sonnet 5.5Xhigh1746 Elo

工具条件未记录 / high / Artificial Analysis independent harness · AA-Briefcase v1.1; Artificial Analysis independent harness; high effort.

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / high / Artificial Analysis independent harness · AA-Briefcase v1.1; Artificial Analysis independent harness; high effort.
模型与配置成绩(Elo)记录与证据
Claude Opus 5.5High1705 Elo

工具条件未记录 / xhigh / Artificial Analysis independent harness · AA-Briefcase v1.1; Artificial Analysis independent harness; xhigh effort.

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / xhigh / Artificial Analysis independent harness · AA-Briefcase v1.1; Artificial Analysis independent harness; xhigh effort.
模型与配置成绩(Elo)记录与证据
Claude Opus 5.5Xhigh1780 Elo

工具条件未记录 / max / Artificial Analysis independent harness · AA-Briefcase v1.1; Artificial Analysis independent harness; max effort.

AA-Briefcase · v1.1 · Elo rating · 工具条件未记录 / max / Artificial Analysis independent harness · AA-Briefcase v1.1; Artificial Analysis independent harness; max effort.
模型与配置成绩(Elo)记录与证据
Claude Opus 5.5Max1822 Elo

Rubric score %

AA-Briefcase v1.1 rubric score;最高可用 effort,逐项未标。

AA-Briefcase · v1.1 · Rubric score · AA-Briefcase v1.1 rubric score;最高可用 effort,逐项未标。
模型与配置成绩(%)记录与证据
Claude Fable 5.2未注明66.1%

历史记录 · 6 次发布

  • Qwen4-Max本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)
  • Claude Fable 5.2本次发布有 3 条成绩记录,含 3 条精确值(可含历史记录)
  • Claude Haiku 5.5本次发布有 2 条成绩记录,含 2 条精确值(可含历史记录)
  • Claude Sonnet 5.5本次发布有 2 条成绩记录,含 2 条精确值(可含历史记录)
  • Claude Opus 5.5本次发布有 3 条成绩记录,含 3 条精确值(可含历史记录)
  • Claude Fable 5.1本次发布有 5 条成绩记录,含 5 条精确值(可含历史记录)