# Claude Opus 4.6 vs GPT-5.4

> Side-by-side benchmark comparison for Claude Opus 4.6 and GPT-5.4 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-opus-4-6-vs-gpt-5-4
- Last updated: October 10, 2026

- Shared sourced benchmarks: 27
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.8

## Quick Verdict

Pick GPT-5.4 if you want the stronger benchmark profile. Claude Opus 4.6 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- GPT-5.4 leads overall 67.03 to 62.12.
- The clearest category separation is in instruction following, where the averages are 52.4 for Claude Opus 4.6 and 90.7 for GPT-5.4.
- The biggest single benchmark swing is HealthBench Hard in Knowledge, with scores of 14.8% and 40.1%.
- GPT-5.4 is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-5.4 also has the larger context window at 1.05M.

## Model Snapshot

| Property | Claude Opus 4.6 | GPT-5.4 |
|----------|----------|----------|
| Creator | Anthropic | OpenAI |
| Type | Proprietary | Proprietary |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 1M | 1.05M |
| Overall Score | 62.12 | 67.03 |
| Benchmarks Covered | 36 | 41 |
| Pricing (input/output) | $5.00 / $25.00 | $2.50 / $15.00 |

## Category Breakdown

### Agentic

- Winner: GPT-5.4
- Claude Opus 4.6 public-lane score: 44.3 (Supported · #54/123)
- GPT-5.4 public-lane score: 50.2 (Supported · #45/123)

| Benchmark | Claude Opus 4.6 | GPT-5.4 | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 65.4% | 75.1% | GPT-5.4 |
| BrowseComp | 83.7% | 82.7% | Claude Opus 4.6 |
| OSWorld-Verified | 72.7% | 75% | GPT-5.4 |
| AndroidWorld | 62.0% | Coming soon | Coming soon |
| Claw-Eval | 70.4% | 60.3% | Claude Opus 4.6 |
| DeepSearchQA | 73.7% | 73.6% | Claude Opus 4.6 |
| CyberGym | 66.6% | 79.0% | GPT-5.4 |
| Gert Labs | 61.85% | 64.89% | GPT-5.4 |
| ResearchClawBench | 19.9% | 15.3% | Claude Opus 4.6 |
| JobBench | 36.7% | 38.9% | GPT-5.4 |
| ApprenticeBench | 5% | 11% | GPT-5.4 |
| MCP Atlas | Coming soon | 70.6% | Coming soon |
| Toolathlon | Coming soon | 54.6% | Coming soon |
| τ²-bench results | Coming soon | 98.9% | Coming soon |
| ExploitGym | Coming soon | 6.0% | Coming soon |

### Coding

- Winner: GPT-5.4
- Claude Opus 4.6 public-lane score: 47 (Supported · #49/146)
- GPT-5.4 public-lane score: 48.1 (Supported · #45/146)

| Benchmark | Claude Opus 4.6 | GPT-5.4 | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 80.8% | Coming soon | Coming soon |
| SWE-bench Verified* | 75.6% | Coming soon | Coming soon |
| LiveCodeBench Pro | 70.7% | 87.5% | GPT-5.4 |
| SWE-bench Pro | 53.4% | 57.7% | GPT-5.4 |
| SWE-Rebench | 65.3% | Coming soon | Coming soon |
| React Native Evals | 84.1% | 85.3% | GPT-5.4 |
| Terminal-Bench 2.0 | 65.4% | 75.1% | GPT-5.4 |
| Vibe Code Bench | 57.57% | 67.42% | GPT-5.4 |
| FrontierCode 1.1 Main | 26.9% | Coming soon | Coming soon |
| PostTrainBench v1.1 | Coming soon | 19.0% | Coming soon |

### Reasoning

- Winner: Not comparable
- Claude Opus 4.6 public-lane score: 72.1 (Unranked · 2 rankable rows)
- GPT-5.4 public-lane score: 64.3 (#23/28)

| Benchmark | Claude Opus 4.6 | GPT-5.4 | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | Coming soon | 74.0% | Coming soon |
| ARC-AGI-3 | Coming soon | 0.2% | Coming soon |
| ARC-AGI-1 | Coming soon | 93.67% | Coming soon |

### Multimodal & Grounded

- Winner: Directional only
- Claude Opus 4.6 public-lane score: 74.5 (#25/54)
- GPT-5.4 public-lane score: 73.1 (#30/54)

| Benchmark | Claude Opus 4.6 | GPT-5.4 | Winner |
|-----------|-----------|-----------|--------|
| MMMU-Pro | 77.3% | 81.2% | GPT-5.4 |
| ERQA | 51.6% | 65.4% | GPT-5.4 |
| ScreenSpot Pro | 83.1% | 85.4% | GPT-5.4 |
| MedXpertQA (MM) | 64.8% | 77.1% | GPT-5.4 |
| OfficeQA Pro | Coming soon | 53.2% | Coming soon |
| MMMU-Pro w/ Python | Coming soon | 82.1% | Coming soon |
| CharXiv | Coming soon | 82.8% | Coming soon |
| SimpleVQA | Coming soon | 61.1% | Coming soon |
| ZeroBench | Coming soon | 41.0% | Coming soon |

### Knowledge

- Winner: Directional only
- Claude Opus 4.6 public-lane score: 58.9 (Estimated · #48/177)
- GPT-5.4 public-lane score: 67.4 (Supported · #22/177)

| Benchmark | Claude Opus 4.6 | GPT-5.4 | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 91.3% | 92.8% | GPT-5.4 |
| GPQA-D | 89.2% | 92.8% | GPT-5.4 |
| SuperGPQA | 95% | Coming soon | Coming soon |
| MMLU-Pro | 82% | Coming soon | Coming soon |
| MMLU-Pro (Arcee) | 89.1% | Coming soon | Coming soon |
| HLE | 53% | 52.1% | Claude Opus 4.6 |
| HLE w/o tools | 40% | 39.8% | Claude Opus 4.6 |
| HealthBench Hard | 14.8% | 40.1% | GPT-5.4 |
| MedXpertQA (Text) | 52.1% | 59.6% | GPT-5.4 |
| HealthBench Professional | Coming soon | 48.1% | Coming soon |

### Instruction Following

- Winner: Directional only
- Claude Opus 4.6 public-lane score: 52.4 (#79/127)
- GPT-5.4 public-lane score: 90.7 (#23/127)

### Mathematics

- Winner: Not comparable
- Claude Opus 4.6 public-lane score: 59 (Unranked · 3 rankable rows)
- GPT-5.4 public-lane score: 64.9 (Unranked · 2 rankable rows)

| Benchmark | Claude Opus 4.6 | GPT-5.4 | Winner |
|-----------|-----------|-----------|--------|
| AIME25 (Arcee) | 99.8% | Coming soon | Coming soon |
| FrontierMath v2 (Tiers 1-3) | 40.700% | 47.600% | GPT-5.4 |
| FrontierMath v2 (Tier 4) | 22.900% | 27.100% | GPT-5.4 |

## FAQ

### Which is better overall, Claude Opus 4.6 or GPT-5.4?

GPT-5.4 is ahead overall on BenchLM right now.

### Where is the biggest gap between Claude Opus 4.6 and GPT-5.4?

The widest category gap is in instruction following, where the averages are 52.4 for Claude Opus 4.6 and 90.7 for GPT-5.4.

### How many benchmarks does Claude Opus 4.6 cover on BenchLM?

Claude Opus 4.6 currently has 36 sourced benchmark scores on BenchLM.

### How many benchmarks does GPT-5.4 cover on BenchLM?

GPT-5.4 currently has 41 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Claude Opus 4.6 vs Claude Opus 4.6 (Adaptive)](/compare/claude-opus-4-6-vs-claude-opus-4-6-thinking)
- [GPT-5.4 vs Claude Opus 4.6 (Adaptive)](/compare/claude-opus-4-6-thinking-vs-gpt-5-4)
- [Claude Opus 4.6 vs GPT-5.4 Pro](/compare/claude-opus-4-6-vs-gpt-5-4-pro)
- [GPT-5.4 vs GPT-5.4 Pro](/compare/gpt-5-4-vs-gpt-5-4-pro)

## Explore More

- [Claude Opus 4.6 profile](/models/claude-opus-4-6)
- [GPT-5.4 profile](/models/gpt-5-4)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)