# GLM-5 Benchmark Scores & Performance

> GLM-5 by Z.AI scores 54.84/100 overall, ranking #69 out of 954 AI models.

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

Last updated: October 17, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | Z.AI |
| Source Type | Open Weight |
| Reasoning Type | Non-Reasoning |
| Context Window | 200K |
| Overall Score | 54.84/100 |
| Overall Rank | #69 of 954 |

## Family & Coverage

- Family: GLM-5
- Variant: base
- Benchmarks covered: 50 of 639
- Sibling models: [GLM-5.3](/models/glm-5-3), [GLM-5.2](/models/glm-5-2), [GLM-5.3-Flash](/models/glm-5-3-flash), [GLM-5.1](/models/glm-5-1), [GLM-5-Turbo](/models/glm-5-turbo), [GLM-5V-Turbo](/models/glm-5v-turbo), [GLM-5 (Reasoning)](/models/glm-5-reasoning)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 56.2% | [Z.AI: GLM-5](https://docs.z.ai/guides/llm/glm-5) |
| [BrowseComp](/benchmarks/browsecomp) | 62% | [GLM-5 model card](https://huggingface.co/zai-org/GLM-5) |
| [HLE w/ tools](/benchmarks/hlewithtools) | 50.4% | [GLM-5 model card](https://huggingface.co/zai-org/GLM-5) |
| [Claw-Eval](/benchmarks/claw-eval) | 57.7% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [QwenClawBench](/benchmarks/qwenclawbench) | 54.1% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [τ³-bench results](/benchmarks/tau3-bench) | 65.6% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [DeepPlanning](/benchmarks/deepplanning) | 14.6% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [Toolathlon](/benchmarks/toolathlon) | 38% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [MCP Atlas](/benchmarks/mcpatlas) | 31.1% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [MCP-Tasks](/benchmarks/mcptasks) | 60.8% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [WideResearch](/benchmarks/wideresearch) | 69.8% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [τ²-bench results](/benchmarks/tau2-bench) | 98.2% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [CyberGym](/benchmarks/cybergym) | 43.2% | [CyberGym leaderboard](https://cybergym.io/) |
| [APEX-Agents-AA](/benchmarks/apexagentsaa) | 14.5% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [Gert Labs](/benchmarks/gertlabs) | 50.99% | [Gert Labs rankings](https://gertlabs.com/rankings) |

## Coding Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [SWE-bench Verified](/benchmarks/swe-bench-verified) | 77.8% | [Z.AI: GLM-5](https://docs.z.ai/guides/llm/glm-5) |
| [SWE-bench Verified*](/benchmarks/swe-bench-verified) | 72.8% | [Arcee Trinity-Large-Thinking comparison table](https://www.arcee.ai/blog/trinity-large-thinking) |
| [SWE-bench Pro](/benchmarks/swe-bench-pro) | 55.1% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [SWE Multilingual](/benchmarks/swe-bench-multilingual) | 73.3% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [NL2Repo](/benchmarks/nl2repo) | 35.9% | [GLM-5.1 model card](https://huggingface.co/zai-org/GLM-5.1) |
| [SWE-Rebench](/benchmarks/swe-rebench) | 62.8% | [SWE-Rebench leaderboard](https://swe-rebench.com/?insight=sep_2025) |
| [React Native Evals](/benchmarks/reactnativeevals) | 74.8% | [React Native Evals leaderboard](https://rn-evals.vercel.app/) |
| [Terminal-Bench 2.0](/benchmarks/terminal-bench-2) | 56.2% | [Z.AI: GLM-5](https://docs.z.ai/guides/llm/glm-5) |

## Multimodal & Grounded Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [Design Arena Website](/benchmarks/designarenawebsite) | 1249 | [OpenRouter model benchmarks](https://openrouter.ai/z-ai/glm-5/benchmarks) |

## Reasoning Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [LongBench v2](/benchmarks/longbench-v2) | 60.8% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [AI-Needle](/benchmarks/aineedle) | 63.3% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [AA-LCR](/benchmarks/lcr) | 75.7% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [CritPt](/benchmarks/critpt) | 2.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |

## Knowledge Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [GPQA](/benchmarks/gpqa) | 86% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [GPQA-D](/benchmarks/gpqa-diamond) | 86.0% | [GLM-5 model card](https://huggingface.co/zai-org/GLM-5) |
| [SuperGPQA](/benchmarks/supergpqa) | 66.8% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [MMLU-Pro](/benchmarks/mmlu-pro) | 85.7% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [MMLU-Pro (Arcee)](/benchmarks/mmlu-pro) | 85.8% | [Arcee Trinity-Large-Thinking comparison table](https://www.arcee.ai/blog/trinity-large-thinking) |
| [HLE](/benchmarks/hle) | 30.5% | [GLM-5 model card](https://huggingface.co/zai-org/GLM-5) |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 27.9% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [AA-GPQA Diamond](/benchmarks/gpqa-diamond) | 82.0% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [AA-HLE](/benchmarks/aahle) | 29.3% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | 0.3% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 26.3% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 35.3% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |

## Instruction Following Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [IFEval](/benchmarks/ifeval) | 92.6% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [AA-IFBench](/benchmarks/ifbench) | 72.3% | [Artificial Analysis model benchmarks](https://artificialanalysis.ai/models/glm-5) |

## Multilingual Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [MMLU-ProX](/benchmarks/mmluprox) | 83.1% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [NOVA-63](/benchmarks/nova63) | 55.1% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |

## Mathematics Benchmarks

| Benchmark | Score | Source |
|-----------|-------|--------|
| [AIME26](/benchmarks/aime2026) | 95.8% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [AIME25 (Arcee)](/benchmarks/aime2025arcee) | 93.3% | [Arcee Trinity-Large-Thinking comparison table](https://www.arcee.ai/blog/trinity-large-thinking) |
| [HMMT Feb 2025](/benchmarks/hmmtfeb2025) | 97.5% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [HMMT Nov 2025](/benchmarks/hmmtnov2025) | 96.9% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [HMMT Feb 2026](/benchmarks/hmmtfeb2026) | 86.4% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |
| [MMAnswerBench](/benchmarks/mmanswerbench) | 82.5% | [Qwen3.6-Plus comparison table](https://qwen.ai/blog?id=qwen3.6) |

## Other Z.AI Models

- [GLM-5.3](/models/glm-5-3) - Score: 67.88
- [GLM-5.2](/models/glm-5-2) - Score: 60.5
- [GLM-5.3-Flash](/models/glm-5-3-flash) - Score: 58.52
- [GLM-5.1](/models/glm-5-1) - Score: 56.31
- [GLM-5-Turbo](/models/glm-5-turbo) - Score: 53.19
- [GLM-5V-Turbo](/models/glm-5v-turbo) - Score: 49.6
- [GLM-4.7](/models/glm-4-7) - Score: 48.85
- [GLM-4.6](/models/glm-4-6) - Score: 40.02
- [GLM-4.7-Flash](/models/glm-4-7-flash) - Score: 39.13
- [GLM-4.5-Air](/models/glm-4-5-air) - Score: 34.93
- [GLM-4.5](/models/glm-4-5) - Score: not computed
- [GLM-5 (Reasoning)](/models/glm-5-reasoning) - Score: not computed