Model comparison
Claude Opus 5 vs GPT-5.6 Luna
Head-to-head evidence from 25 shared benchmark results across 5 categories. Overall scores shown here use the public BenchAlign v5 ranking lane.
Public leaderboard positions: Claude Opus 5 #1 (Estimated); GPT-5.6 Luna #24 (Estimated). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Claude Opus 5 and GPT-5.6 Luna share 25 comparable benchmark results. 5 of 8 categories are comparable. 54 results are unique to Claude Opus 5; 17 to GPT-5.6 Luna.
Updated July 24, 2026- Shared results
- 25
- Claude Opus 5 only
- 54
- GPT-5.6 Luna only
- 17
- Comparable categories
- 5 / 8
Pick Claude Opus 5 if you want the stronger benchmark profile. GPT-5.6 Luna only becomes the better choice if knowledge is the priority or you want the cheaper token bill.
Confidence note. This is a partial-evidence comparison with 25 shared benchmark results across 5 evidence categories; 5 of 8 categories currently have scoreable aggregates for both models. Treat the verdict as directional until coverage is more balanced.
Why this result
Claude Opus 5 is clearly ahead on the BenchAlign aggregate, 85.88 to 66.59. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Claude Opus 5's sharpest advantage is in reasoning, where it averages 90.4 against 59.5. The single biggest benchmark swing on the page is ARC-AGI-2, 90.4% to 59.5%. GPT-5.6 Luna does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
Claude Opus 5 is also the more expensive model on tokens at $5.00 input / $25.00 output per 1M tokens, versus $1.00 input / $6.00 output per 1M tokens for GPT-5.6 Luna. That is roughly 4.2x on output cost alone.
Category breakdown
Exact category averages are shown below. Not measured means BenchLM does not have enough sourced public coverage for that model and category.
| Category | Claude Opus 5 | Δ | GPT-5.6 Luna |
|---|---|---|---|
| Reasoning | Claude Opus 590.4 | Margin← 30.9 | GPT-5.6 Luna59.5 |
| Knowledge | Claude Opus 564.7 | Margin→ 27.6 | GPT-5.6 Luna92.3 |
| Coding | Claude Opus 589.5 | Margin← 26.8 | GPT-5.6 Luna62.7 |
| Multimodal | Claude Opus 566.9 | Margin→ 11.5 | GPT-5.6 Luna78.4 |
| Agentic | Claude Opus 590.8 | Margin← 6.7 | GPT-5.6 Luna84.1 |
| Math | Claude Opus 5Not measured | MarginNo overlap | GPT-5.6 Luna73.6 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
ARC-AGI-2
ReasoningA 90.4%B 59.5%Winner: Claude Opus 5Δ 30.9ARC-AGI-2: Claude Opus 5 scored 90.4%; GPT-5.6 Luna scored 59.5%. Claude Opus 5 wins this benchmark. - Source ↗
SWE-bench Pro
CodingA 79.2%B 62.7%Winner: Claude Opus 5Δ 16.5SWE-bench Pro: Claude Opus 5 scored 79.2%; GPT-5.6 Luna scored 62.7%. Claude Opus 5 wins this benchmark. - Source ↗
BrowseComp
AgenticA 90.8%B 83.3%Winner: Claude Opus 5Δ 7.5BrowseComp: Claude Opus 5 scored 90.8%; GPT-5.6 Luna scored 83.3%. Claude Opus 5 wins this benchmark.
Operational comparison
Runtime and commercial metrics are compared only when both models have a complete sourced value.
| Metric | Claude Opus 5 | GPT-5.6 Luna | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Claude Opus 5$5 input / $25 output | GPT-5.6 Luna$1 input / $6 output | GPT-5.6 Luna has the lower combined listed price. |
| Generation speedtokens per second | Claude Opus 5Not available | GPT-5.6 LunaNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Claude Opus 5Not available | GPT-5.6 LunaNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Claude Opus 5Not available | GPT-5.6 Luna1M | A complete context comparison is not available. |
Benchmark Deep Dive
AgenticClaude Opus 5 wins32 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Luna | Result |
|---|---|---|---|
| FrontierBench v0.1Source | 43.3% | — | Not comparable |
| BrowseCompSource | 90.8% | 83.3% | Claude Opus 5 leads |
| HLE w/ toolsSource | 64.7% | — | Not comparable |
| DeepSearchQASource | 95.0% | — | Not comparable |
| DRACOSource | 88.6% | — | Not comparable |
| BrowseComp (10-agent, prerelease)Source | 93.6% | — | Not comparable |
| OSWorld 2.0Source | 70.6% | 45.6% | Claude Opus 5 leads |
| MCP AtlasSource | 85.8% | — | Not comparable |
| MCP-Atlas claim coverageSource | 89.1% | — | Not comparable |
| LAB all-pass (Anthropic harness)Source | 23.58% | — | Not comparable |
| LAB criterion-pass (Anthropic harness)Source | 93.74% | — | Not comparable |
| LAB all-pass (Harvey held-out)Source | 11.7% | — | Not comparable |
| LAB criterion-pass (Harvey held-out)Source | 94.1% | — | Not comparable |
| GDPval-AASource | 1861 | 1582 | Claude Opus 5 leads |
| Toolathlon-VerifiedSource | 80.6% | — | Not comparable |
| Toolathlon Verified Pass@3Source | 87.0% | — | Not comparable |
| Toolathlon Verified Pass³Source | 73.1% | — | Not comparable |
| Toolathlon Verified avg. turnsSource | 23.5 turns | — | Not comparable |
| AutomationBenchSource | 26.0% | — | Not comparable |
| AA Agentic IndexSource | 55.3% | 45.6% | Claude Opus 5 leads |
| GDPval-AASource | 68.0% | 54.1% | Claude Opus 5 leads |
| AA Tau3 BankingSource | 30.3% | 27.2% | Claude Opus 5 leads |
| AA BriefcaseSource | 1720 | — | Not comparable |
| aaTerminalBench21Source | 89.1% | 80.9% | Claude Opus 5 leads |
| Terminal-Bench 2.0Source | — | 84.7% | Not comparable |
| CyberGymSource | — | 77.9% | Not comparable |
| ExploitGymSource | — | 12.4% | Not comparable |
| ToolathlonSource | — | 53.4% | Not comparable |
| AA Harvey LABSource | — | 87.9% | Not comparable |
| AA ITBenchSource | — | 40.3% | Not comparable |
| AA AutomationBenchSource | — | 42.2% | Not comparable |
| APEX-Agents-AASource | — | 35.8% | Not comparable |
CodingClaude Opus 5 wins13 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Luna | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 96% | — | Not comparable |
| SWE-bench ProSource | 79.2% | 62.7% | Claude Opus 5 leads |
| SWE MultilingualSource | 89.5% | — | Not comparable |
| SWE MultimodalSource | 59.4% | — | Not comparable |
| deepSweSource | 68.8% | 67.2% | Claude Opus 5 leads |
| FrontierCode 1.1 MainSource | 53.4% | — | Not comparable |
| FrontierCode 1.1 ExtendedSource | 63.6% | 55.1% | Claude Opus 5 leads |
| ProgramBench (episode 1)Source | 83.0% | — | Not comparable |
| ProgramBenchSource | 93.0% | — | Not comparable |
| cursorBench32Source | 70.0% | 61.1% | Claude Opus 5 leads |
| AA Coding IndexSource | 78.0% | 71.5% | Claude Opus 5 leads |
| AA-SciCodeSource | 55.7% | 52.5% | Claude Opus 5 leads |
| Terminal-Bench 2.0Source | — | 84.7% | Not comparable |
ReasoningClaude Opus 5 wins5 benchmarks
KnowledgeGPT-5.6 Luna wins24 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Luna | Result |
|---|---|---|---|
| HLESource | 64.7% | — | Not comparable |
| HLE w/o toolsSource | 56.3% | — | Not comparable |
| HealthBench (raw)Source | 67.1% | — | Not comparable |
| HealthBench (length-adjusted)Source | 57.8% | — | Not comparable |
| HealthBench ProfessionalSource | 59.8% | 55.7% | Claude Opus 5 leads |
| HealthBench Professional (raw)Source | 73.4% | — | Not comparable |
| BioMysteryBench (human-solvable)Source | 90.1% | — | Not comparable |
| BioMysteryBench (human-difficult)Source | 49.4% | — | Not comparable |
| SpatialBench VerifiedSource | 72.5% | — | Not comparable |
| SingleCellBenchSource | 60.6% | — | Not comparable |
| ProteinGym HardSource | 47.7% | — | Not comparable |
| Protein DesignSource | 42.5% | — | Not comparable |
| Organic chemistry V2Source | 61.6% | — | Not comparable |
| Protocols (troubleshooting)Source | 61.1% | — | Not comparable |
| Protocols (understanding)Source | 78.4% | — | Not comparable |
| Artificial Analysis Intelligence IndexSource | 60.7% | 51.2% | Claude Opus 5 leads |
| AA-GPQA DiamondSource | 93.2% | 91.1% | Claude Opus 5 leads |
| AA-HLESource | 52.6% | 37.2% | Claude Opus 5 leads |
| AA-Omniscience IndexSource | 31.3% | -11.2% | Claude Opus 5 leads |
| AA-Omniscience AccuracySource | 54.2% | 41.5% | Claude Opus 5 leads |
| AA-Omniscience Hallucination RateSource | 50.1% | 90.1% | Claude Opus 5 leads |
| GPQASource | — | 92.3% | Not comparable |
| GPQA-DSource | — | 92.3% | Not comparable |
| HealthBench HardSource | — | 32.0% | Not comparable |
Math8 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Luna | Result |
|---|---|---|---|
| IMO 2026Source | 42/42 | — | Not comparable |
| RiemannBench (no tools)Source | 60.0% | — | Not comparable |
| RiemannBench (tools)Source | 79.0% | — | Not comparable |
| ArXivMath Jun. 2026 (no tools)Source | 90.8% | — | Not comparable |
| ArXivMath Jun. 2026 (tools)Source | 91.3% | — | Not comparable |
| FrontierMath (legacy)Source | — | 78.6% | Not comparable |
| FrontierMath v2 (Tiers 1-3)Source | — | 78.600% | Not comparable |
| FrontierMath v2 (Tier 4)Source | — | 58.500% | Not comparable |
Multilingual3 benchmarks
MultimodalGPT-5.6 Luna wins11 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Luna | Result |
|---|---|---|---|
| Chartography (no tools)Source | 29.6% | — | Not comparable |
| Chartography (tools)Source | 83.0% | — | Not comparable |
| BenchCAD Vision2Code (no tools)Source | 0.366 | — | Not comparable |
| BenchCAD Vision2Code (tools)Source | 0.821 | — | Not comparable |
| GDP.pdf (no tools)Source | 83.4% | — | Not comparable |
| GDP.pdf (tools)Source | 85.5% | — | Not comparable |
| OfficeQASource | 78.1% | — | Not comparable |
| OfficeQA ProSource | 66.9% | — | Not comparable |
| AA-MMMU-ProSource | 84.7% | 78.6% | Claude Opus 5 leads |
| MMMU-ProSource | — | 78.4% | Not comparable |
| MMMU-Pro w/ PythonSource | — | 79.5% | Not comparable |
Frequently Asked Questions (6)
Which is better, Claude Opus 5 or GPT-5.6 Luna?
Claude Opus 5 is ahead on BenchLM's BenchAlign leaderboard, 85.88 to 66.59. The biggest single separator in this matchup is ARC-AGI-2, where the scores are 90.4% and 59.5%.
Which is better for knowledge tasks, Claude Opus 5 or GPT-5.6 Luna?
GPT-5.6 Luna has the edge for knowledge tasks in this comparison, averaging 92.3 versus 64.7. Inside this category, AA-Omniscience Index is the benchmark that creates the most daylight between them.
Which is better for coding, Claude Opus 5 or GPT-5.6 Luna?
Claude Opus 5 has the edge for coding in this comparison, averaging 89.5 versus 62.7. Inside this category, SWE-bench Pro is the benchmark that creates the most daylight between them.
Which is better for reasoning, Claude Opus 5 or GPT-5.6 Luna?
Claude Opus 5 has the edge for reasoning in this comparison, averaging 90.4 versus 59.5. Inside this category, ARC-AGI-2 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Claude Opus 5 or GPT-5.6 Luna?
Claude Opus 5 has the edge for agentic tasks in this comparison, averaging 90.8 versus 84.1. Inside this category, GDPval-AA is the benchmark that creates the most daylight between them.
Which is better for multimodal and grounded tasks, Claude Opus 5 or GPT-5.6 Luna?
GPT-5.6 Luna has the edge for multimodal and grounded tasks in this comparison, averaging 78.4 versus 66.9. Inside this category, AA-MMMU-Pro is the benchmark that creates the most daylight between them.