Zhipu (GLM) · model
GLM-5 benchmark scores
As of 2026-10-07, GLM-5 (Zhipu (GLM)) has measured scores on 42 benchmarks and estimated scores on 66 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 82.0% | measured |
| AA-HLE | 29.3% | measured |
| AA-SciCode | 47.8% | estimated ± 3.8 pp, high confidence |
| AIME26 | 95.8% | measured |
| Artificial Analysis Intelligence Index | 27.9% | measured |
| CritPt | 2.0% | measured |
| GDPval-AA | 21.6% | estimated ± 3.7 pp, low confidence |
| GPQA Diamond | 86.0% | measured |
| HLE | 50.4% | measured |
| IFBench | 70.8% | estimated ± 6.9 pp, medium confidence |
| AA-LCR | 75.7% | measured |
| LiveCodeBench v6 | 86.9% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 26.3% | measured |
| SciCode | 47.5% | estimated ± 3.8 pp, high confidence |
| BrowseComp | 73.4% | estimated ± 3.4 pp, high confidence |
| HLE w/ tools | 35.7% | estimated ± 2.4 pp, medium confidence |
| IFEval | 92.6% | measured |
| LongBench v2 | 60.8% | measured |
| VITA-Bench | 38.8% | estimated ± 8.2 pp, low confidence |
| τ²-bench results | 98.2% | measured |
| AA Coding Index | 56.0% | estimated ± 5.1 pp, medium confidence |
| APEX-Agents | 0.0% | estimated ± 0.6 pp, low confidence |
| APEX-Agents-AA | 14.5% | measured |
| BioMysteryBench (human-difficult) | 32.3% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 76.8% | estimated ± 11.5 pp, low confidence |
| SWE-bench Verified | 77.8% | measured |
| AutomationBench | 28.4% | estimated ± 5.8 pp, low confidence |
| BFCL v4 | 49.3% | estimated ± 8.6 pp, low confidence |
| CyberGym | 43.2% | measured |
| JobBench | 0.0% | estimated ± 5.1 pp, low confidence |
| SkillsBench | 36.3% | estimated ± 5.6 pp, low confidence |
| SWE-bench Pro | 55.1% | measured |
| τ³-bench results | 65.6% | measured |
| WideResearch | 69.8% | measured |
| AA Agentic Index | 17.7% | estimated ± 5.1 pp, medium confidence |
| MMLU-Pro | 85.7% | measured |
| FrontierMath v2 (Tier 4) | 2.1% | measured |
| FrontierMath v2 (Tiers 1-3) | 16.4% | measured |
| GPQA | 86.0% | measured |
| AA-IFBench | 72.3% | measured |
| Gert Labs | 51.0% | measured |
| AA EnterpriseOps-Gym | 52.2% | estimated ± 3.6 pp, medium confidence |
| ApprenticeBench | 0.0% | estimated ± 3.1 pp, low confidence |
| ARC-AGI-1 | 82.4% | estimated ± 2.7 pp, medium confidence |
| ARC-AGI-2 | 42.4% | estimated ± 10.1 pp, low confidence |
| CursorBench 3.1 | 51.3% | estimated ± 5.6 pp, low confidence |
| CursorBench 3.2 | 50.6% | estimated ± 3.6 pp, high confidence |
| FrontierCode 1.1 Main | 18.0% | estimated ± 2.7 pp, low confidence |
| FrontierSWE v2 | 6.4% | estimated ± 9.0 pp, medium confidence |
| OSWorld-Verified | 65.6% | estimated ± 4.7 pp, medium confidence |
| Vals GPQA Diamond | 85.1% | estimated ± 2.0 pp, high confidence |
| Vals LiveCodeBench | 82.4% | estimated ± 3.8 pp, low confidence |
| Vals MMLU-Pro | 83.2% | estimated ± 1.2 pp, high confidence |
| Vals SWE-bench | 76.0% | estimated ± 1.4 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 49.2% | estimated ± 6.3 pp, medium confidence |
| AA AutomationBench | 0.0% | estimated ± 13.0 pp, low confidence |
| AA ITBench | 13.2% | estimated ± 3.6 pp, low confidence |
| AA Tau3 Banking | 4.4% | estimated ± 11.8 pp, low confidence |
| CursorBench 4.0 | 31.7% | estimated ± 3.6 pp, medium confidence |
| HLE w/o tools | 27.0% | estimated ± 2.9 pp, high confidence |
| OSWorld 2.0 | 0.4% | estimated ± 4.5 pp, low confidence |
| PostTrainBench v1.1 | 22.8% | estimated ± 4.7 pp, medium confidence |
| SWE Multilingual | 73.3% | measured |
| Toolathlon-Verified | 61.2% | estimated ± 2.2 pp, low confidence |
| Vibe Code Bench | 27.7% | estimated ± 6.5 pp, low confidence |
| ExploitGym | 1.5% | estimated ± 9.3 pp, low confidence |
| C-Eval | 91.3% | estimated ± 0.7 pp, medium confidence |
| Claw-Eval | 57.7% | measured |
| DeepPlanning | 14.6% | measured |
| HMMT Feb 2025 | 97.5% | measured |
| HMMT Feb 2026 | 86.4% | measured |
| HMMT Nov 2025 | 96.9% | measured |
| MCP Atlas | 31.1% | measured |
| MCP-Tasks | 60.8% | measured |
| MMAnswerBench | 82.5% | measured |
| MMLU-ProX | 83.1% | measured |
| MMLU-Redux | 93.4% | estimated ± 0.7 pp, high confidence |
| NL2Repo | 42.8% | estimated ± 6.1 pp, medium confidence |
| NOVA-63 | 55.1% | measured |
| QwenClawBench | 54.1% | measured |
| SuperGPQA | 66.8% | measured |
| Terminal-Bench 2.0 | 56.2% | measured |
| Toolathlon | 38.0% | measured |
| HealthBench Hard | 0.0% | estimated ± 5.1 pp, low confidence |
| LiveCodeBench Pro | 79.8% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (Text) | 43.7% | estimated ± 4.5 pp, low confidence |
| MMLU-Pro (Arcee) | 85.8% | measured |
| React Native Evals | 74.8% | measured |
| ResearchClawBench | 15.8% | estimated ± 1.3 pp, high confidence |
| SWE-Rebench | 62.8% | measured |
| ARC-AGI-3 | 0.0% | estimated ± 2.1 pp, medium confidence |
| FrontierMath (legacy) | 17.4% | estimated ± 0.7 pp, low confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| FrontierCode 1.1 Extended | 54.3% | estimated ± 3.0 pp, low confidence |
| HealthBench (raw) | 38.5% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 42.4% | estimated ± 4.8 pp, low confidence |
| HealthBench Professional (raw) | 49.6% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 65.4% | estimated ± 1.8 pp, low confidence |
| AIME 2025 | 96.4% | estimated ± 1.5 pp, medium confidence |
| LiveCodeBench | 85.1% | estimated ± 10.2 pp, low confidence |
| Agents' Last Exam | 24.0% | estimated ± 2.1 pp, medium confidence |
| OpenHarmony Bench | 53.3% | estimated ± 4.9 pp, low confidence |
| IMOAnswerBench | 81.7% | estimated ± 5.9 pp, medium confidence |
| MMMLU | 88.2% | estimated ± 1.1 pp, medium confidence |
| MMLU | 89.5% | estimated ± 1.1 pp, medium confidence |
| MATH-500 | 97.7% | estimated ± 3.1 pp, medium confidence |