Home Model series GLM GLM-5.3 GLM-5.3 · Terminal-Bench · 2.1;GLM 官方图仅列数值,具体 score / mean reward 指标未注明 · Scores and configurations GLM-5.3 · Terminal-Bench · 2.1;GLM 官方图仅列数值,具体 score / mean reward 指标未注明 · Scores and configurations GLM-5.3 record for Terminal-Bench · 2.1;GLM 官方图仅列数值,具体 score / mean reward 指标未注明. 1 score record for this release, including 1 exact value (may include historical records). Includes 2 source locations.
Scores and evaluation conditions GLM-5.3 · Terminal-Bench scores and configurations Metric and score Configuration and evaluation conditions Sources Reported value; metric and unit unspecified 88.2 Reported 88.2 · Unit not specified Published Unconfirmed
GLM-5.3 官方表格配置未详 Settings 官方模型卡 GLM-5.3 表格成绩。卡片建议 benchmark reproduction 使用默认 reasoning_effort=max,但未逐项确认所有表格成绩均以该配置测得,故不将其套用到未注明项。 Unknowns 各未逐项注明的 benchmark reasoning_effort 未知、评测使用官方 API 还是本地权重未注明、精度/量化与 serving 版本未注明 Tools Tool access not recorded Reasoning budget Output limit 65536 tokens Framework Claude Code 2.1.207 Protocol Claude Code 2.1.207; temperature=1.0; top_p=1; max_new_tokens=65536; 6-hour timeout. Chart/table does not identify score versus mean reward. Conditions GLM-5.3; exact metric, unit, effort, endpoint and precision unspecified. View this score in rankings
Benchmark versions and metrics Aliases GLM-5.3-Flash Terminal Bench 2.1 chart value
Reported value; metric and unit unspecified · Unit not specified · Ranking direction unconfirmed