# Claude Sonnet 4.6 vs GPT-5.6 Luna

> Side-by-side benchmark comparison for Claude Sonnet 4.6 and GPT-5.6 Luna across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-sonnet-4-6-vs-gpt-5-6-luna
- Last updated: September 24, 2026

- Shared sourced benchmarks: 10
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.7

## Quick Verdict

Pick GPT-5.6 Luna if you want the stronger benchmark profile. Claude Sonnet 4.6 only makes more sense when its price, context window, or workload-specific category wins matter more than the overall score.

## Summary

- GPT-5.6 Luna leads overall 65.6 to 56.35.
- The clearest category separation is in mathematics, where the averages are 48.8 for Claude Sonnet 4.6 and 94.2 for GPT-5.6 Luna.
- The biggest single benchmark swing is FrontierMath v2 (Tier 4) in Mathematics, with scores of 8.300% and 58.500%.
- GPT-5.6 Luna is the cheaper option on output tokens, which matters if you expect large responses or heavy interactive use.
- GPT-5.6 Luna also has the larger context window at 1.05M.

## Model Snapshot

| Property | Claude Sonnet 4.6 | GPT-5.6 Luna |
|----------|----------|----------|
| Creator | Anthropic | OpenAI |
| Type | Proprietary | Proprietary |
| Reasoning | Non-Reasoning | Reasoning |
| Context | 200K | 1.05M |
| Overall Score | 56.35 | 65.6 |
| Benchmarks Covered | 26 | 29 |
| Pricing (input/output) | $3.00 / $15.00 | $0.20 / $1.20 |

## Category Breakdown

### Agentic

- Winner: GPT-5.6 Luna
- Claude Sonnet 4.6 public-lane score: 41.8 (Supported · #44/105)
- GPT-5.6 Luna public-lane score: 55.3 (Supported · #28/105)

| Benchmark | Claude Sonnet 4.6 | GPT-5.6 Luna | Winner |
|-----------|-----------|-----------|--------|
| Terminal-Bench 2.0 | 59.1% | Coming soon | Coming soon |
| OSWorld-Verified | 72.1% | Coming soon | Coming soon |
| Claw-Eval | 67.8% | Coming soon | Coming soon |
| CyberGym | 65.2% | 77.9% | GPT-5.6 Luna |
| Gert Labs | 62.92% | Coming soon | Coming soon |
| OSWorld 2.0 | 8.3% | 45.6% | GPT-5.6 Luna |
| JobBench | 36.9% | Coming soon | Coming soon |
| Terminal-Bench 2.1 (Vals) | 57.3% | 79.0% | GPT-5.6 Luna |
| ApprenticeBench | 2% | 7% | GPT-5.6 Luna |
| Terminal-Bench 3.0 | Coming soon | 14.3% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 84.7% | Coming soon |
| BrowseComp | Coming soon | 83.3% | Coming soon |
| ExploitGym | Coming soon | 12.4% | Coming soon |
| Toolathlon | Coming soon | 53.4% | Coming soon |

### Coding

- Winner: GPT-5.6 Luna
- Claude Sonnet 4.6 public-lane score: 48.1 (Supported · #41/135)
- GPT-5.6 Luna public-lane score: 64.5 (Supported · #9/135)

| Benchmark | Claude Sonnet 4.6 | GPT-5.6 Luna | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 79.6% | Coming soon | Coming soon |
| SWE-Rebench | 60.7% | Coming soon | Coming soon |
| React Native Evals | 80.6% | Coming soon | Coming soon |
| Vibe Code Bench | 51.48% | Coming soon | Coming soon |
| cursorBench31 | 48.8% | Coming soon | Coming soon |
| FrontierCode 1.1 Main | 24.3% | Coming soon | Coming soon |
| LiveCodeBench (Vals) | 82.1% | Coming soon | Coming soon |
| SWE-bench (Vals) | 77.4% | 93.0% | GPT-5.6 Luna |
| SWE-bench Pro | Coming soon | 62.7% | Coming soon |
| Terminal-Bench 2.1 | Coming soon | 84.7% | Coming soon |
| DeepSWE | Coming soon | 67.2% | Coming soon |
| FrontierCode 1.1 Extended | Coming soon | 55.1% | Coming soon |
| cursorBench32 | Coming soon | 61.1% | Coming soon |
| VulcanBench v3 | Coming soon | 85.5% | Coming soon |

### Reasoning

- Winner: Not comparable
- Claude Sonnet 4.6 public-lane score: 69 (Unranked · 2 rankable rows)
- GPT-5.6 Luna public-lane score: 54.7 (#18/19)

| Benchmark | Claude Sonnet 4.6 | GPT-5.6 Luna | Winner |
|-----------|-----------|-----------|--------|
| ARC-AGI-2 | Coming soon | 59.5% | Coming soon |
| ARC-AGI-3 | Coming soon | 0.2% | Coming soon |

### Multimodal & Grounded

- Winner: Directional only
- Claude Sonnet 4.6 public-lane score: 54.1 (#35/50)
- GPT-5.6 Luna public-lane score: 67.1 (#22/50)

| Benchmark | Claude Sonnet 4.6 | GPT-5.6 Luna | Winner |
|-----------|-----------|-----------|--------|
| CharXiv | 77.4% | Coming soon | Coming soon |
| MMMU-Pro | Coming soon | 78.4% | Coming soon |
| MMMU-Pro w/ Python | Coming soon | 79.5% | Coming soon |

### Knowledge

- Winner: GPT-5.6 Luna
- Claude Sonnet 4.6 public-lane score: 53 (Supported · #48/158)
- GPT-5.6 Luna public-lane score: 64.6 (Supported · #22/158)

| Benchmark | Claude Sonnet 4.6 | GPT-5.6 Luna | Winner |
|-----------|-----------|-----------|--------|
| GPQA | 89.9% | 92.3% | GPT-5.6 Luna |
| SuperGPQA | 95% | Coming soon | Coming soon |
| MMLU-Pro | 79.2% | Coming soon | Coming soon |
| HLE | 49% | Coming soon | Coming soon |
| GPQA Diamond (Vals) | 85.6% | 91.7% | GPT-5.6 Luna |
| MMLU-Pro (Vals) | 87.3% | 86.0% | Claude Sonnet 4.6 |
| GPQA-D | Coming soon | 92.3% | Coming soon |
| HealthBench Professional | Coming soon | 55.7% | Coming soon |
| HealthBench Hard | Coming soon | 32.0% | Coming soon |

### Instruction Following

- Winner: Not comparable
- Claude Sonnet 4.6 public-lane score: 46.6 (#88/124)
- GPT-5.6 Luna public-lane score: Coming soon

### Mathematics

- Winner: Not comparable
- Claude Sonnet 4.6 public-lane score: 48.8 (Unranked · 2 rankable rows)
- GPT-5.6 Luna public-lane score: 94.2 (Unranked · 3 rankable rows)

| Benchmark | Claude Sonnet 4.6 | GPT-5.6 Luna | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | 32.400% | 78.600% | GPT-5.6 Luna |
| FrontierMath v2 (Tier 4) | 8.300% | 58.500% | GPT-5.6 Luna |
| FrontierMath (legacy) | Coming soon | 78.6% | Coming soon |

## FAQ

### Which is better overall, Claude Sonnet 4.6 or GPT-5.6 Luna?

GPT-5.6 Luna is ahead overall on BenchLM right now.

### Where is the biggest gap between Claude Sonnet 4.6 and GPT-5.6 Luna?

The widest category gap is in mathematics, where the averages are 48.8 for Claude Sonnet 4.6 and 94.2 for GPT-5.6 Luna.

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

Claude Sonnet 4.6 currently has 26 sourced benchmark scores on BenchLM.

### How many benchmarks does GPT-5.6 Luna cover on BenchLM?

GPT-5.6 Luna currently has 29 sourced benchmark scores on BenchLM.

## Related Comparisons

- [Claude Sonnet 4.6 vs GPT-5.6 Sol](/compare/claude-sonnet-4-6-vs-gpt-5-6-sol)
- [GPT-5.6 Luna vs GPT-5.6 Sol](/compare/gpt-5-6-luna-vs-gpt-5-6-sol)
- [Claude Sonnet 4.6 vs GPT-5.6 Terra](/compare/claude-sonnet-4-6-vs-gpt-5-6-terra)
- [GPT-5.6 Luna vs GPT-5.6 Terra](/compare/gpt-5-6-luna-vs-gpt-5-6-terra)

## Explore More

- [Claude Sonnet 4.6 profile](/models/claude-sonnet-4-6)
- [GPT-5.6 Luna profile](/models/gpt-5-6-luna)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)