Home Model series GLM GLM-5 GLM-5 · Humanity's Last Exam · GLM-5固定v2报告/模型卡;text-only · w/Tools · Scores and configurations GLM-5 · Humanity's Last Exam · GLM-5固定v2报告/模型卡;text-only · w/Tools · Scores and configurations GLM-5 record for Humanity's Last Exam · GLM-5固定v2报告/模型卡;text-only · w/Tools. 1 score record for this release, including 1 exact value (may include historical records). Includes 2 source locations.
Scores and evaluation conditions GLM-5 · Humanity's Last Exam scores and configurations Metric and score Configuration and evaluation conditions Sources Reported score 50.4 Reported 50.4 · Unit not specified Published Unconfirmed
GLM-5 · 报告与固定模型卡 Settings 最终后训练GLM-5;各行scaffold、解码与工具条件分别见protocol,未将工具或来源另立为推理档位。 Unknowns 来源未逐项明确思考开关、reasoning effort、服务精度与完整题集修订。、Tool-Decathlon报告39.2与固定卡38.0均保留;BrowseComp上下文管理的正文与附录方法描述不一致。 Tools Tools allowed Reasoning budget Output limit 131072 tokens Framework not recorded Protocol HLE with tools; maximum context length 202,752 tokens; GPT-5.2 (medium) judge. Thinking mode and reasoning effort are not stated. Conditions HLE with tools; maximum context length 202,752 tokens; GPT-5.2 (medium) judge. Thinking mode and reasoning effort are not stated. View this score in rankings
Benchmark versions and metrics Aliases Humanity's Last Exam
Reported score · Unit not specified · Ranking direction unconfirmed