GLM-5.3
开始使用
API 接入
Zhipu AI API文档
目录部署标识：glm-5.3。具体区域、权限与价格以服务商为准。
相关动态
订阅更新公开成绩与计算依据
综合参考分根据已公开的评测资料整理。查看发布日期依据
综合参考分 59.6 · 19项基准
推理求解 63.4
GPQA Diamond · 0.917171717171717 比例
Artificial Analysis · 独立评测 · Diamond
GLM-5.3 (Max) · Current AA standalone protocol. Status verified in aa-method-current.md section gpqa-diamond
Terminal-Bench-Science · 5.71%
Vals AI · 独立评测 · 0.1.0
GLM-5.3 / Mini-SWE-agent · Current displayed evaluator row. Mini-SWE-agent. Scores are not pooled across harnesses; source methodology and repeat counts retained.
HLE · 0.179 比例
Artificial Analysis · 独立评测 · Diamond
GLM-5.3 (Max) · Artificial Analysis current published protocol; hle-diamond
CritPt · 0.335 比例
Artificial Analysis · 独立评测 · 2 Hard
GLM-5.3 (Max) · Artificial Analysis current published protocol; critpt-v2-hard
工具与任务执行 72.7
APEX-Agents · 38.1 %
huggingface.co · 官方发布比较 · current published task suite
GLM-5.3 / official Hy4 comparison, APEX-Agents pass@1 · Original numeric labels visually read in official image; APEX-Agents; Official Hy4 preview blue series; not Hy3 cyan sub-series. JPEG read at decoder 1/8 scale, original SHA-256 preserved.
tau3-Banking · 0.503092783505155 比例
Artificial Analysis · 独立评测 · 1.0.1
GLM-5.3 (Max) · Current AA standalone protocol. Status verified in aa-method-current.md section tau3-banking
GDPval-AA · 1656.88 分
Artificial Analysis · 独立评测 · 2.1
GLM-5.3 (Max) · Artificial Analysis current published protocol; gdpval-aa-normalized
Terminal-Bench · 38.89%
Vals AI · 独立评测 · 4.0
GLM-5.3 / Mini-SWE-agent · Current displayed evaluator row. Mini-SWE-agent. Scores are not pooled across harnesses; source methodology and repeat counts retained.
AutomationBench · 0.622028649962642 比例
Artificial Analysis · 独立评测 · current published task suite
GLM-5.3 (Max) · Artificial Analysis current published protocol; automationbench-aa
AA-Briefcase · 1502.84 分
Artificial Analysis · 独立评测 · 1.1
GLM-5.3 (Max) · Artificial Analysis current published protocol; aa-briefcase
EnterpriseOps-Gym · 0.36436884512085943 比例
Artificial Analysis · 独立评测 · current published task suite
GLM-5.3 (Max) · Artificial Analysis current published protocol; enterprise-ops-gym-aa
AA-AnalystAgent · 0.3875 比例
Artificial Analysis · 独立评测 · current AA pass^5
GLM-5.3 (Max) · Artificial Analysis current published protocol; aa-analyst-agent
编程 58.6
FrontierCode · 0.21587095959596 比例
Artificial Analysis · 独立评测 · 2.0
GLM-5.3 (Max) · Artificial Analysis current published protocol; frontiercode-2-0
Vals Code Migration · 55.07%
Vals AI · 独立评测 · 1.1
GLM-5.3 / Published Code Migration 1.1 sandbox · Current displayed evaluator row. Published Code Migration 1.1 sandbox. Scores are not pooled across harnesses; source methodology and repeat counts retained.
DeepSWE · 68.1
腾讯 Hy4 官方发布比较表 · 官方发布比较 · 1.1
腾讯官方比较表 · GLM-5.3 · DeepSWE current 1.1 task suite; mini-swe-agent; sandbox 8 CPUs and 16 GB RAM
语言与信息处理 39.6
GDP.pdf · 0.112 比例
Artificial Analysis · 独立评测 · current published task suite
GLM-5.3 (Max) · Artificial Analysis current published protocol; gdp-pdf
SimpleQA Verified · 0.41 比例
Epoch AI · 独立评测 · run 1.2.0
glm-5.3_max · epoch-simpleqa-reviewed-method-20261029-task-version-1.2.0 · 2026-08-28公开
AA-Omniscience · 0.3233333333333333 比例
Artificial Analysis · 独立评测 · 1.1
GLM-5.3 (Max) · Artificial Analysis current published protocol; omniscience-accuracy
AA-Omniscience · 0.6113300492610838 比例
Artificial Analysis · 独立评测 · 1.1
GLM-5.3 (Max) · Artificial Analysis current published protocol; omniscience-non-hallucination
资料来源
- 版本日期来源 · 版本资料 · 2026-10-29查阅
- Artificial Analysis · independent_evaluation · 2026-10-29查阅
- Vals AI · independent_evaluation · 2026-10-29查阅
- huggingface.co · official_release_comparison · 2026-10-29查阅
- Vals AI · independent_evaluation · 2026-10-29查阅
- Scale AI · independent_evaluation · 2026-10-29查阅
- Vals AI · independent_evaluation · 2026-10-29查阅
- Epoch AI · independent_evaluation · 2026-10-29查阅