benchgap
Zhipu (GLM) · model

GLM-5.2 benchmark scores

As of 2026-11-22, GLM-5.2 (Zhipu (GLM)) has measured scores on 53 benchmarks and estimated scores on 115 more.

BenchmarkScoreSource
AA-GPQA Diamond89.5%measured
AA-HLE41.1%measured
AA-SciCode51.2%measured
AIME2699.2%measured
Artificial Analysis Intelligence Index33.7%measured
CritPt20.9%measured
GDPval-AA43.6%measured
GPQA Diamond91.2%measured
HLE40.5%measured
IFBench71.3%estimated ± 6.6 pp, medium confidence
AA-LCR78.3%measured
LiveCodeBench v691.6%estimated ± 1.7 pp, high confidence
AA-Omniscience Accuracy24.3%measured
SciCode51.0%estimated ± 3.1 pp, high confidence
Terminal-Bench 2.181.0%measured
BrowseComp85.3%estimated ± 4.2 pp, high confidence
HLE w/ tools54.7%measured
IFEval92.9%estimated ± 2.8 pp, high confidence
LongBench v264.2%estimated ± 5.5 pp, medium confidence
VITA-Bench46.9%estimated ± 9.9 pp, low confidence
τ²-bench results99.1%measured
AA Coding Index68.8%measured
APEX-Agents34.3%estimated ± 4.9 pp, low confidence
APEX-Agents-AA33.7%measured
BioMysteryBench (human-difficult)48.3%estimated ± 6.8 pp, low confidence
DeepSearchQA91.6%estimated ± 4.5 pp, high confidence
SWE-bench Verified80.7%estimated ± 2.3 pp, high confidence
BFCL v476.1%estimated ± 2.7 pp, medium confidence
CyberGym75.8%estimated ± 5.3 pp, medium confidence
JobBench35.3%estimated ± 7.3 pp, low confidence
SkillsBench56.4%estimated ± 9.1 pp, low confidence
SWE-bench Pro62.1%measured
WideResearch80.2%estimated ± 3.3 pp, high confidence
AA Agentic Index39.4%measured
MMLU-Pro85.6%estimated ± 2.5 pp, low confidence
AA-IFBench73.3%measured
FailureSensorIQ66.9%estimated ± 1.7 pp, high confidence
GPQA91.2%measured
MMLU90.5%estimated ± 0.9 pp, high confidence
Vals GPQA Diamond85.6%measured
Vals LiveCodeBench69.5%measured
Vals MMLU-Pro86.7%measured
MATH-50098.2%estimated ± 3.1 pp, low confidence
MMMLU90.9%estimated ± 0.7 pp, low confidence
Chess Puzzles22.0%measured
AIME 2025100.0%estimated ± 1.8 pp, low confidence
AA MMLU-Pro85.0%estimated ± 0.7 pp, high confidence
Gert Labs49.1%measured
AA-AnalystAgent46.5%estimated ± 3.7 pp, medium confidence
AA EnterpriseOps-Gym42.7%measured
AA Harvey LAB v1.092.0%estimated ± 1.2 pp, medium confidence
AA Tau3 Banking39.6%estimated ± 2.0 pp, medium confidence
AA Terminal-Bench 2.182.1%estimated ± 2.5 pp, high confidence
ApprenticeBench10.0%estimated ± 5.0 pp, low confidence
ARC-AGI-191.1%estimated ± 2.5 pp, high confidence
ARC-AGI-263.5%estimated ± 9.7 pp, medium confidence
BabaIsBench96.0%measured
CursorBench 3.147.0%estimated ± 3.4 pp, low confidence
CursorBench 3.255.0%measured
EBR-bench9.5%measured
Terminal-Bench 3.04.6%measured
FrontierCode 1.1 Main37.9%estimated ± 1.6 pp, low confidence
FrontierMath v2 (Tier 4)29.3%measured
FrontierMath v2 (Tiers 1-3)59.2%measured
FrontierSWE v219.2%estimated ± 8.3 pp, medium confidence
Furniture Assembly24.0%estimated ± 7.4 pp, medium confidence
MaxxBench33.1%estimated ± 4.3 pp, low confidence
MazeBench0.0%measured
MirrorCode36.7%measured
Mystery Game Puzzles15.0%measured
NYT Connections74.3%measured
OSWorld-Verified77.7%estimated ± 5.1 pp, medium confidence
Terminal-Bench 4.10.0%estimated ± 4.8 pp, medium confidence
Terminal-Bench Hard50.8%measured
Vals SWE-bench82.8%measured
Terminal-Bench 2.1 (Vals AI)67.8%measured
VulcanBench v376.7%estimated ± 5.1 pp, low confidence
CWE-Bench-AA55.8%estimated ± 5.2 pp, medium confidence
AA Cyber Index24.2%estimated ± 9.2 pp, low confidence
DeepsecBench-AA16.9%estimated ± 8.2 pp, medium confidence
HealthBench Professional (length-adjusted, Artificial Analysis)56.1%estimated ± 4.3 pp, low confidence
HealthBench Professional (raw, Artificial Analysis)66.0%estimated ± 0.8 pp, medium confidence
AA ITBench42.7%measured
Bug Hunt Bench24.6%estimated ± 6.3 pp, low confidence
CursorBench 4.030.3%estimated ± 2.0 pp, medium confidence
CWE-bench v157.4%estimated ± 5.1 pp, medium confidence
DeepSWE72.9%estimated ± 3.8 pp, medium confidence
DRACO86.9%estimated ± 0.7 pp, low confidence
FrontierCode 2.016.9%estimated ± 5.4 pp, low confidence
FrontierPhysics2.5%estimated ± 5.0 pp, low confidence
HLE w/o tools40.5%measured
OSWorld 2.016.1%estimated ± 6.9 pp, low confidence
OSWorld 2.161.1%estimated ± 3.6 pp, high confidence
OSWorld 2.1 Offline40.2%estimated ± 1.9 pp, low confidence
PostTrainBench v1.131.7%measured
ProgramBench63.7%measured
SpireClimb5.8%estimated ± 7.2 pp, medium confidence
StealthMark48.0%estimated ± 8.7 pp, low confidence
Toolathlon-Verified67.6%estimated ± 2.3 pp, medium confidence
VulcanBench Frontier v488.5%estimated ± 1.2 pp, medium confidence
AA AutomationBench41.9%estimated ± 7.9 pp, low confidence
GDP.pdf19.6%estimated ± 4.2 pp, high confidence
AA Terminal-Bench 4.04.3%estimated ± 11.0 pp, low confidence
ARC-AGI-33.5%estimated ± 7.2 pp, medium confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, medium confidence
cursorBench4144.8%estimated ± 3.7 pp, medium confidence
CWE-bench46.4%estimated ± 6.2 pp, medium confidence
MKA35.5%estimated ± 4.6 pp, high confidence
SkillsBench official leaderboard (BenchFlow)51.5%estimated ± 6.1 pp, medium confidence
BinaryAudit35.8%estimated ± 12.1 pp, low confidence
CompileBench73.3%estimated ± 10.0 pp, low confidence
OTelBench17.5%estimated ± 4.1 pp, low confidence
Vibe Code Bench50.5%estimated ± 10.0 pp, low confidence
FACTS Multimodal42.1%estimated ± 4.0 pp, high confidence
Hermes Bench70.0%estimated ± 1.8 pp, medium confidence
Hermes Index37.2%estimated ± 3.1 pp, high confidence
ExploitBench28.8%estimated ± 7.2 pp, low confidence
ExploitGym11.2%estimated ± 2.7 pp, high confidence
C-Eval95.7%estimated ± 1.0 pp, low confidence
Claw-Eval73.3%estimated ± 3.2 pp, medium confidence
DeepPlanning48.5%estimated ± 13.1 pp, low confidence
HMMT Feb 202596.3%estimated ± 1.9 pp, high confidence
HMMT Feb 202692.5%measured
HMMT Nov 202594.4%measured
MCP Atlas76.8%measured
MCP-Tasks75.6%estimated ± 7.0 pp, low confidence
MMAnswerBench91.0%measured
MMLU-Redux94.0%estimated ± 0.8 pp, medium confidence
NL2Repo48.9%measured
PostTrainBench v134.3%measured
QwenClawBench63.6%estimated ± 2.6 pp, medium confidence
SuperGPQA80.3%estimated ± 7.2 pp, medium confidence
Terminal-Bench 2.067.8%estimated ± 5.4 pp, medium confidence
Toolathlon48.2%measured
AA LiveCodeBench93.3%estimated ± 2.6 pp, medium confidence
HealthBench Hard14.3%estimated ± 5.1 pp, low confidence
LiveCodeBench Pro85.3%estimated ± 6.0 pp, low confidence
MedXpertQA (Text)57.5%estimated ± 4.5 pp, medium confidence
MMLU-Pro (Arcee)87.3%estimated ± 3.4 pp, low confidence
React Native Evals81.4%estimated ± 3.2 pp, high confidence
ResearchClawBench20.7%measured
PostTrainBench30.4%measured
DAYJOB1.3%estimated ± 3.2 pp, high confidence
DAYJOB Healthcare0.7%estimated ± 4.2 pp, high confidence
Finance Agent v255.7%estimated ± 2.4 pp, medium confidence
BioMysteryBench (human-solvable)89.2%estimated ± 0.6 pp, medium confidence
FrontierCode 1.1 Extended55.1%estimated ± 1.6 pp, low confidence
HealthBench (raw)49.7%estimated ± 6.0 pp, medium confidence
HealthBench (length-adjusted)54.0%estimated ± 3.4 pp, medium confidence
HealthBench Professional50.1%estimated ± 4.2 pp, medium confidence
HealthBench Professional (raw)60.3%estimated ± 4.5 pp, high confidence
HLE-Verified16.4%estimated ± 2.0 pp, low confidence
LABBench281.3%estimated ± 1.8 pp, low confidence
ScarfBench62.9%estimated ± 5.5 pp, low confidence
BlenderWorld (text-only)70.0%estimated ± 2.8 pp, medium confidence
Cookie-Bench76.9%estimated ± 1.3 pp, medium confidence
STEMBench2.4%estimated ± 0.2 pp, low confidence
SimpleQA24.0%estimated ± 4.3 pp, medium confidence
Agents' Last Exam31.3%estimated ± 9.7 pp, medium confidence
OpenHarmony Bench v1.058.4%measured
SEC-Bench Pro21.1%estimated ± 7.5 pp, low confidence
IMOAnswerBench84.6%estimated ± 5.6 pp, medium confidence
MRCRv285.5%estimated ± 11.3 pp, low confidence
MMLongBench-Doc61.3%estimated ± 2.6 pp, low confidence
SWE-Marathon56.0%estimated ± 12.9 pp, low confidence
BioBasicsBench25.4%estimated ± 2.2 pp, low confidence
LifeVal75.2%estimated ± 3.8 pp, medium confidence
ExploitBench Internal Port0.0%estimated ± 1.4 pp, low confidence