# Claude Opus 4.5 vs GLM-5

> Side-by-side benchmark comparison for Claude Opus 4.5 and GLM-5 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-opus-4-5-vs-glm-5
- Last updated: September 18, 2026

- Shared sourced benchmarks: 30
- HTML indexing: indexable
- Ranking lane: BenchAlign v5

## Quick Verdict

Pick GLM-5 if you want the stronger benchmark profile. Claude Opus 4.5 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- GLM-5 leads overall 61.47 to 58.42.
- The clearest category separation is in instruction following, where the averages are 30.6 for Claude Opus 4.5 and 87.2 for GLM-5.
- The biggest single benchmark swing is HLE in Knowledge, with scores of 30.8% and 50.4%.
- GLM-5 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.

## Model Snapshot

| Property | Claude Opus 4.5 | GLM-5 |
|----------|----------|----------|
| Creator | Anthropic | Z.AI |
| Type | Proprietary | Open Weight |
| Reasoning | Non-Reasoning | Non-Reasoning |
| Context | 200K | 200K |
| Overall Score | 58.42 | 61.47 |
| Benchmarks Covered | 45 | 36 |
| Pricing (input/output) | $5.00 / $25.00 | $1.00 / $3.20 |

## Category Breakdown

### Agentic

- Winner: Coming soon
- Claude Opus 4.5 public-lane score: 43.2 (Estimated · #102/154)
- GLM-5 public-lane score: 51 (Estimated · #48/154)

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 59.3% | 56.2% | Claude Opus 4.5 |
| OSWorld-Verified | 66.3% | Coming soon | Coming soon |
| OSWorld | 66.3% | Coming soon | Coming soon |
| Claw-Eval | 59.6% | 57.7% | Claude Opus 4.5 |
| QwenClawBench | 52.3% | 54.1% | GLM-5 |
| τ³-bench results | 70.2% | 65.6% | Claude Opus 4.5 |
| VITA-Bench | 23.3% | Coming soon | Coming soon |
| DeepPlanning | 26.4% | 14.6% | Claude Opus 4.5 |
| Toolathlon | 43.5% | 38% | Claude Opus 4.5 |
| MCP Atlas | 42.3% | 31.1% | Claude Opus 4.5 |
| MCP-Tasks | 71.8% | 60.8% | Claude Opus 4.5 |
| WideResearch | 76.4% | 69.8% | Claude Opus 4.5 |
| CyberGym | 50.6% | 43.2% | Claude Opus 4.5 |
| Gert Labs | 64.23% | 50.99% | Claude Opus 4.5 |
| JobBench | 32.3% | Coming soon | Coming soon |

### Coding

- Winner: Coming soon
- Claude Opus 4.5 public-lane score: 56.7 (Estimated · #36/154)
- GLM-5 public-lane score: 56 (Estimated · #39/154)

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 80.9% | 77.8% | Claude Opus 4.5 |
| LiveCodeBench v6 | 84.8% | Coming soon | Coming soon |
| SWE-bench Pro | 57.1% | 55.1% | Claude Opus 4.5 |
| SWE Multilingual | 77.5% | 73.3% | Claude Opus 4.5 |
| NL2Repo | 43.2% | Coming soon | Coming soon |
| SWE-bench Verified* | Coming soon | 72.8% | Coming soon |
| SWE-Rebench | Coming soon | 62.8% | Coming soon |
| React Native Evals | Coming soon | 74.8% | Coming soon |

### Multimodal & Grounded

- Winner: Coming soon
- Claude Opus 4.5 public-lane score: 23.5 (#46/48)
- GLM-5 public-lane score: Coming soon

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 70.6% | Coming soon | Coming soon |
| MathVision | 74.3% | Coming soon | Coming soon |
| CharXiv | 68.5% | Coming soon | Coming soon |
| VideoMMMU | 84.4% | Coming soon | Coming soon |
| ScreenSpot Pro | 45.7% | Coming soon | Coming soon |
| V* | 67.0% | Coming soon | Coming soon |

### Reasoning

- Winner: Coming soon
- Claude Opus 4.5 public-lane score: 72.1 (Unranked · 4 rankable rows)
- GLM-5 public-lane score: 52.1 (Unranked · 4 rankable rows)

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| LongBench v2 | 64.4% | 60.8% | Claude Opus 4.5 |
| AI-Needle | 74% | 63.3% | Claude Opus 4.5 |

### Knowledge

- Winner: Coming soon
- Claude Opus 4.5 public-lane score: 54 (Estimated · #60/184)
- GLM-5 public-lane score: 54.1 (Estimated · #58/184)

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 87% | 86% | Claude Opus 4.5 |
| SuperGPQA | 70.6% | 66.8% | Claude Opus 4.5 |
| MMLU-Pro | 89.5% | 85.7% | Claude Opus 4.5 |
| MMLU-Redux | 96.6% | Coming soon | Coming soon |
| C-Eval | 92.2% | Coming soon | Coming soon |
| HLE | 30.8% | 50.4% | GLM-5 |
| GPQA-D | Coming soon | 86.0% | Coming soon |
| MMLU-Pro (Arcee) | Coming soon | 85.8% | Coming soon |

### Instruction Following

- Winner: Coming soon
- Claude Opus 4.5 public-lane score: 30.6 (#114/124)
- GLM-5 public-lane score: 87.2 (#32/124)

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| IFEval | 90.9% | 92.6% | GLM-5 |
| IFBench | 58% | Coming soon | Coming soon |

### Multilingual

- Winner: Claude Opus 4.5
- Claude Opus 4.5 public-lane score: 82.9 (#2/12)
- GLM-5 public-lane score: 48.7 (#6/12)

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| MMLU-ProX | 85.7% | 83.1% | Claude Opus 4.5 |
| NOVA-63 | 56.7% | 55.1% | Claude Opus 4.5 |

### Mathematics

- Winner: Claude Opus 4.5
- Claude Opus 4.5 public-lane score: 57.9 (#6/7)
- GLM-5 public-lane score: 56.5 (#7/7)

| Benchmark | Claude Opus 4.5 | GLM-5 | Winner |
|-----------|-----------|-----------|--------|
| AIME26 | 95.1% | 95.8% | GLM-5 |
| HMMT Feb 2025 | 92.9% | 97.5% | GLM-5 |
| HMMT Nov 2025 | 93.3% | 96.9% | GLM-5 |
| HMMT Feb 2026 | 85.3% | 86.4% | GLM-5 |
| MMAnswerBench | 84.0% | 82.5% | Claude Opus 4.5 |
| FrontierMath v2 (Tiers 1-3) | 20.690% | 16.434% | Claude Opus 4.5 |
| FrontierMath v2 (Tier 4) | 4.167% | 2.100% | Claude Opus 4.5 |
| AIME25 (Arcee) | Coming soon | 93.3% | Coming soon |

## FAQ

### Which is better overall, Claude Opus 4.5 or GLM-5?

GLM-5 is ahead overall on BenchLM right now.

### Where is the biggest gap between Claude Opus 4.5 and GLM-5?

The widest category gap is in instruction following, where the averages are 30.6 for Claude Opus 4.5 and 87.2 for GLM-5.

### How many benchmarks does Claude Opus 4.5 cover on BenchLM?

Claude Opus 4.5 currently has 45 sourced benchmark scores on BenchLM.

### How many benchmarks does GLM-5 cover on BenchLM?

GLM-5 currently has 36 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Claude Opus 4.5 vs Claude Opus 4.5 Thinking](/compare/claude-opus-4-5-vs-claude-opus-4-5-thinking)
- [GLM-5 vs Claude Opus 4.5 Thinking](/compare/claude-opus-4-5-thinking-vs-glm-5)
- [Claude Opus 4.5 vs GLM-5.3](/compare/claude-opus-4-5-vs-glm-5-3)
- [GLM-5 vs GLM-5.3](/compare/glm-5-vs-glm-5-3)

## Explore More

- [Claude Opus 4.5 profile](/models/claude-opus-4-5)
- [GLM-5 profile](/models/glm-5)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)
