GLM-5.3 · Terminal-Bench(GLM-5.3模型卡;3.0) 28.3 · 主文证据(ev-gmc-glm-5-3-76)

GLM-5.3 的 Terminal-Bench · GLM-5.3模型卡;3.0 官方出处。GLM-5.3 · max · 28.3。见《GLM-5.3 official model card》Benchmark table;Terminal Bench 3.0;GLM-5.3目标列。证据 ev-gmc-glm-5-3-76。

原文与定位

GLM-5.3 · Terminal-Bench(GLM-5.3模型卡;3.0),GLM-5.3 · max配置,Reported score 28.3(具体成绩公布时间未确认)。见《GLM-5.3 official model card》Benchmark table;Terminal Bench 3.0;GLM-5.3目标列。 查看成绩条件

| Benchmark | GLM-5.3 | GLM-5.2 | Kimi K3 | DeepSeek-V4 Pro-0813 | Qwen3.8-Max | Opus 4.8 | Fable 5 (w/ fallback) | GPT-5.6 Sol | | Terminal Bench 3.0 | 28.3 | 4.6 | 17.4 | – | – | 21.1 | 33.7 | **34.6** | - **Terminal-Bench 3.0**: We evaluate Terminal-Bench-3 tasks with the Claude Code 2.1.207 harness (reasoning effort=max, 400K context, and 128K maximum output), reporting avg@3 over three rollouts per task. Each rollout runs in an isolated container built from the task's official image, and is capped at 600 agent turns with a 10-hour timeout. Tool Search is disabled, and the artifacts each agent produces are scored by the task's official separate verifier. GLM-5.3 supports controlling the thinking budget through the `reasoning_effort` parameter, which accepts three levels: `low`, `high`, and `max`. It defaults to `max` if not passed (or if set to any other value). To use `low` or `high`, pass them explicitly. For benchmark and leaderboard reproduction, keep the default `max`.
官方材料
GLM-5.3 official model card
定位
主文 · Benchmark table;Terminal Bench 3.0;GLM-5.3目标列
材料版本
本地2026-11-17官方卡片快照;2026-11-18完整目标列、内嵌图和原文脚注核对完成。
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

GLM-5.3 · Terminal-Bench(GLM-5.3模型卡;3.0),GLM-5.3 · max配置,Reported score 28.3(具体成绩公布时间未确认)。官方出处 GLM-5.3 official model card(主文,本地2026-11-17官方卡片快照;2026-11-18完整目标列、内嵌图和原文脚注核对完成。)Benchmark table;Terminal Bench 3.0;GLM-5.3目标列,https://huggingface.co/zai-org/GLM-5.3/blob/main/README.md。Benchlog https://benchlog.lekshome.cc/evidence/glm-5.3/ev-gmc-glm-5-3-76/

关联成绩与条件

GLM-5.3 · Terminal-Bench 成绩与配置
指标与成绩配置与评测条件证据关系
Reported score28.3原文 28.3 · 单位 未标注
公布 未确认
GLM-5.3 · max
设置
官方模型卡目标列;逐行scaffold、工具、采样及预算条件见protocol。来源把benchmark复现明确绑定max或行内明确reasoning_effort=max。
未知项
未逐项说明的指标单位、题集修订、服务精度和运行次数保持未知。
工具
工具条件未记录
推理预算
max · 输出上限 128K tokens
框架
Claude Code 2.1.207
协议
The card says reasoning_effort defaults to max and recommends keeping that default for benchmark reproduction. **Terminal-Bench 3.0**: We evaluate Terminal-Bench-3 tasks with the Claude Code 2.1.207 harness (reasoning effort=max, 400K context, and 128K maximum output), reporting avg@3 over three rollouts per task. Each rollout runs in an isolated container built from the task's official image, and is capped at 600 agent turns with a 10-hour timeout. Tool Search is disabled, and the artifacts each agent produces are scored by the task's official separate verifier.
条件
GLM-5.3; exact serving/API endpoint and weight precision not stated. 2026-11-18按完整模型卡修正单位、指标/题集或协议标注,数字不变。

直接支持该成绩

在排名中查看此成绩