HLE w/ Tools · GLM-5.3模型卡;w/ Tools; full/text-only subset not stated · 测试定义与成绩

HLE w/ Tools · GLM-5.3模型卡;w/ Tools; full/text-only subset not stated 收录 1 次模型发布的评测披露或成绩。

测试版本与指标

别名 HLE w/ Tools

  • Reported score · 单位未标注 · 排序方向未确认

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。

Reported score

允许工具 / 推理预算未记录 / 框架未记录 · The card says reasoning_effort defaults to max and recommends keeping that default for benchmark reproduction. **HLE w/ tools**: We use sampling parameters of `temperature=1.0` and `top_p=0.95` for evaluation, with a maximum generation length of `163,840` tokens. The evaluation is conducted with a maximum context length of `300,000` tokens, using a context management strategy. We use GPT-5.6-luna (medium) as the judge model.

HLE w/ Tools · GLM-5.3模型卡;w/ Tools; full/text-only subset not stated · Reported score · 允许工具 / 推理预算未记录 / 框架未记录 · The card says reasoning_effort defaults to max and recommends keeping that default for benchmark reproduction. **HLE w/ tools**: We use sampling parameters of `temperature=1.0` and `top_p=0.95` for evaluation, with a maximum generation length of `163,840` tokens. The evaluation is conducted with a maximum context length of `300,000` tokens, using a context management strategy. We use GPT-5.6-luna (medium) as the judge model.
模型与配置成绩记录与证据
GLM-5.3Max62.5

历史记录 · 1 次发布

  • GLM-5.3本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)