# GLM-5.2 Benchmark Scores & Performance

> GLM-5.2 by Z.AI scores 68.19/100 overall, ranking #26 out of 484 AI models.

Canonical page: https://benchlm.ai/models/glm-5-2

Last updated: September 14, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | Z.AI |
| Source Type | Open Weight |
| Reasoning Type | Reasoning |
| Context Window | 1M |
| Overall Score | 68.19/100 |
| Overall Rank | #26 of 484 |

## Family & Coverage

- Family: GLM-5
- Variant: flagship (5.2)
- Benchmarks covered: 42 of 435
- Sibling models: [GLM-5.3](/models/glm-5-3), [GLM-5.3-Flash](/models/glm-5-3-flash), [GLM-5.1](/models/glm-5-1), [GLM-5-Turbo](/models/glm-5-turbo), [GLM-5](/models/glm-5), [GLM-5V-Turbo](/models/glm-5v-turbo), [GLM-5 (Reasoning)](/models/glm-5-reasoning)
- Related earlier model: [GLM-5.1](/models/glm-5-1)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Terminal-Bench 3.0](/benchmarks/terminal-bench-3) | 4.6% |
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 81% |
| [MCP Atlas](/benchmarks/mcpatlas) | 76.8% |
| [Toolathlon](/benchmarks/toolathlon) | 48.2% |
| [τ²-bench results](/benchmarks/tau2-bench) | 99.1% |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 45.9% |
| [APEX-Agents-AA](/benchmarks/apexagentsaa) | 33.7% |
| [ResearchClawBench](/benchmarks/researchclawbench) | 20.7% |
| [AA ITBench](/benchmarks/aaitbench) | 42.7% |
| [Terminal-Bench 2.1 (Vals)](/benchmarks/valsterminalbench21) | 67.8% |
| [AA Agentic Index](/benchmarks/aaagenticindex) | 39.4% |
| [GDPval-AA](/benchmarks/gdpvalaa) | 1418 |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [SWE-bench Pro](/benchmarks/swe-bench-pro) | 62.1% |
| [NL2Repo](/benchmarks/nl2repo) | 48.9% |
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 81.0% |
| [ProgramBench](/benchmarks/programbench) | 63.7% |
| [CursorBench 3.2](/benchmarks/cursorbench32) | 55.0% |
| [AA-SciCode](/benchmarks/aascicode) | 51.2% |
| [OpenHarmony Bench](/benchmarks/openharmonybench) | 58.4% |
| [LiveCodeBench (Vals)](/benchmarks/valslivecodebench) | 69.5% |
| [SWE-bench (Vals)](/benchmarks/valsswebench) | 82.8% |
| [AA Coding Index](/benchmarks/aacodingindex) | 68.8% |

## Multimodal & Grounded Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Design Arena Website](/benchmarks/designarenawebsite) | 1307 |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [CritPt](/benchmarks/critpt) | 20.9% |
| [AA-LCR](/benchmarks/lcr) | 78.3% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [GPQA](/benchmarks/gpqa) | 91.2% |
| [GPQA-D](/benchmarks/gpqa-diamond) | 91.2% |
| [HLE](/benchmarks/hle) | 54.7% |
| [HLE w/o tools](/benchmarks/hlenotools) | 40.5% |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 38.6% |
| [AA-GPQA Diamond](/benchmarks/aagpqadiamond) | 89.5% |
| [AA-HLE](/benchmarks/aahle) | 41.1% |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | 4.4% |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 24.3% |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 26.3% |
| [GPQA Diamond (Vals)](/benchmarks/valsgpqadiamond) | 85.6% |
| [MMLU-Pro (Vals)](/benchmarks/valsmmlupro) | 86.7% |

## Instruction Following Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-IFBench](/benchmarks/aaifbench) | 73.3% |

## Mathematics Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AIME26](/benchmarks/aime2026) | 99.2% |
| [HMMT Nov 2025](/benchmarks/hmmtnov2025) | 94.4% |
| [HMMT Feb 2026](/benchmarks/hmmtfeb2026) | 92.5% |
| [MMAnswerBench](/benchmarks/mmanswerbench) | 91.0% |

## Other Z.AI Models

- [GLM-5.3](/models/glm-5-3) - Score: 68.44
- [GLM-5.3-Flash](/models/glm-5-3-flash) - Score: 66.06
- [GLM-5.1](/models/glm-5-1) - Score: 63.3
- [GLM-5-Turbo](/models/glm-5-turbo) - Score: 61.72
- [GLM-5](/models/glm-5) - Score: 61.51
- [GLM-4.7](/models/glm-4-7) - Score: 57.78
- [GLM-5V-Turbo](/models/glm-5v-turbo) - Score: 57.72
- [GLM-5 (Reasoning)](/models/glm-5-reasoning) - Score: 57.57
- [GLM-4.5](/models/glm-4-5) - Score: 55.4
- [GLM-4.6](/models/glm-4-6) - Score: 51.9
- [GLM-4.7-Flash](/models/glm-4-7-flash) - Score: 46.35
- [GLM-4.5-Air](/models/glm-4-5-air) - Score: 43.9