Model comparison
Claude Opus 5 vs GPT-5.6 Sol
Head-to-head evidence from 26 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 Sol #4 (Supported). Intervals and evidence labels describe ranking uncertainty, not a guarantee for a specific workload.
Evidence parity. Claude Opus 5 and GPT-5.6 Sol share 26 comparable benchmark results. 5 of 8 categories are comparable. 53 results are unique to Claude Opus 5; 22 to GPT-5.6 Sol.
Updated July 24, 2026- Shared results
- 26
- Claude Opus 5 only
- 53
- GPT-5.6 Sol only
- 22
- Comparable categories
- 5 / 8
Pick Claude Opus 5 if you want the stronger benchmark profile. GPT-5.6 Sol only becomes the better choice if knowledge is the priority.
Confidence note. This is a partial-evidence comparison with 26 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 81.46. 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 coding, where it averages 89.5 against 64.6. The single biggest benchmark swing on the page is SWE-bench Pro, 79.2% to 64.6%. GPT-5.6 Sol does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.
GPT-5.6 Sol is also the more expensive model on tokens at $5.00 input / $30.00 output per 1M tokens, versus $5.00 input / $25.00 output per 1M tokens for Claude Opus 5.
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 Sol |
|---|---|---|---|
| Knowledge | Claude Opus 564.7 | Margin→ 29.9 | GPT-5.6 Sol94.6 |
| Coding | Claude Opus 589.5 | Margin← 24.9 | GPT-5.6 Sol64.6 |
| Multimodal | Claude Opus 566.9 | Margin→ 16.1 | GPT-5.6 Sol83.0 |
| Reasoning | Claude Opus 590.4 | Margin→ 2.1 | GPT-5.6 Sol92.5 |
| Agentic | Claude Opus 590.8 | Margin→ 1.2 | GPT-5.6 Sol92.0 |
| Math | Claude Opus 5Not measured | MarginNo overlap | GPT-5.6 Sol87.5 |
Decisive benchmark drivers
The largest measured benchmark gaps in this matchup, with exact reported values.
More
- Source ↗
SWE-bench Pro
CodingA 79.2%B 64.6%Winner: Claude Opus 5Δ 14.6SWE-bench Pro: Claude Opus 5 scored 79.2%; GPT-5.6 Sol scored 64.6%. Claude Opus 5 wins this benchmark. - Source ↗
ARC-AGI-2
ReasoningA 90.4%B 92.5%Winner: GPT-5.6 SolΔ 2.1ARC-AGI-2: Claude Opus 5 scored 90.4%; GPT-5.6 Sol scored 92.5%. GPT-5.6 Sol wins this benchmark. - Source ↗
BrowseComp
AgenticA 90.8%B 92.2%Winner: GPT-5.6 SolΔ 1.4BrowseComp: Claude Opus 5 scored 90.8%; GPT-5.6 Sol scored 92.2%. GPT-5.6 Sol 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 Sol | Comparison |
|---|---|---|---|
| Input / output priceUSD per 1M tokens | Claude Opus 5$5 input / $25 output | GPT-5.6 Sol$5 input / $30 output | Claude Opus 5 has the lower combined listed price. |
| Generation speedtokens per second | Claude Opus 5Not available | GPT-5.6 SolNot available | A complete speed comparison is not available. |
| First-answer latencyseconds to first token | Claude Opus 5Not available | GPT-5.6 SolNot available | A complete latency comparison is not available. |
| Context windowmaximum listed tokens | Claude Opus 5Not available | GPT-5.6 Sol1M | A complete context comparison is not available. |
Benchmark Deep Dive
AgenticGPT-5.6 Sol wins34 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Result |
|---|---|---|---|
| FrontierBench v0.1Source | 43.3% | — | Not comparable |
| BrowseCompSource | 90.8% | 92.2% | GPT-5.6 Sol 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% | 62.6% | 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 | 1736 | 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% | 54.0% | Claude Opus 5 leads |
| GDPval-AASource | 68.0% | 61.8% | Claude Opus 5 leads |
| AA Tau3 BankingSource | 30.3% | 33.0% | GPT-5.6 Sol leads |
| AA BriefcaseSource | 1720 | 1505 | Claude Opus 5 leads |
| aaTerminalBench21Source | 89.1% | 88% | Claude Opus 5 leads |
| Terminal-Bench 2.0Source | — | 91.9% | Not comparable |
| CyberGymSource | — | 84.5% | Not comparable |
| ExploitGymSource | — | 33.7% | Not comparable |
| ToolathlonSource | — | 58% | Not comparable |
| τ²-bench resultsSource | — | 85.1% | Not comparable |
| AA ITBenchSource | — | 56.2% | Not comparable |
| AA AutomationBenchSource | — | 51.2% | Not comparable |
| AA Harvey LABSource | — | 87.2% | Not comparable |
| terminalBenchHardSource | — | 65.9% | Not comparable |
| AA EnterpriseOps-GymSource | — | 42.9% | Not comparable |
CodingClaude Opus 5 wins14 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Result |
|---|---|---|---|
| SWE-bench VerifiedSource | 96% | — | Not comparable |
| SWE-bench ProSource | 79.2% | 64.6% | Claude Opus 5 leads |
| SWE MultilingualSource | 89.5% | — | Not comparable |
| SWE MultimodalSource | 59.4% | — | Not comparable |
| deepSweSource | 68.8% | 72.7% | GPT-5.6 Sol leads |
| FrontierCode 1.1 MainSource | 53.4% | — | Not comparable |
| FrontierCode 1.1 ExtendedSource | 63.6% | 60.6% | Claude Opus 5 leads |
| ProgramBench (episode 1)Source | 83.0% | — | Not comparable |
| ProgramBenchSource | 93.0% | — | Not comparable |
| cursorBench32Source | 70.0% | 67.2% | Claude Opus 5 leads |
| AA Coding IndexSource | 78.0% | 77.4% | Claude Opus 5 leads |
| AA-SciCodeSource | 55.7% | 56.1% | GPT-5.6 Sol leads |
| Terminal-Bench 2.0Source | — | 91.9% | Not comparable |
| VulcanBench v3Source | — | 87.0% | Not comparable |
ReasoningGPT-5.6 Sol wins6 benchmarks
KnowledgeGPT-5.6 Sol wins24 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Sol | 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% | 60.5% | GPT-5.6 Sol 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% | 58.9% | Claude Opus 5 leads |
| AA-GPQA DiamondSource | 93.2% | 94.1% | GPT-5.6 Sol leads |
| AA-HLESource | 52.6% | 47.2% | Claude Opus 5 leads |
| AA-Omniscience IndexSource | 31.3% | 21.7% | Claude Opus 5 leads |
| AA-Omniscience AccuracySource | 54.2% | 58.5% | GPT-5.6 Sol leads |
| AA-Omniscience Hallucination RateSource | 50.1% | 88.8% | Claude Opus 5 leads |
| GPQASource | — | 94.6% | Not comparable |
| GPQA-DSource | — | 94.6% | Not comparable |
| HealthBench HardSource | — | 33.1% | Not comparable |
Math8 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Sol | 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 | — | 89% | Not comparable |
| FrontierMath v2 (Tiers 1-3)Source | — | 89.000% | Not comparable |
| FrontierMath v2 (Tier 4)Source | — | 83.000% | Not comparable |
Multilingual3 benchmarks
MultimodalGPT-5.6 Sol wins11 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Sol | 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% | 83.4% | Claude Opus 5 leads |
| MMMU-ProSource | — | 83% | Not comparable |
| MMMU-Pro w/ PythonSource | — | 84.6% | Not comparable |
Inst. Following1 benchmarks
| Benchmark | Claude Opus 5 | GPT-5.6 Sol | Result |
|---|---|---|---|
| AA-IFBenchSource | — | 72.7% | Not comparable |
Frequently Asked Questions (6)
Which is better, Claude Opus 5 or GPT-5.6 Sol?
Claude Opus 5 is ahead on BenchLM's BenchAlign leaderboard, 85.88 to 81.46. The biggest single separator in this matchup is SWE-bench Pro, where the scores are 79.2% and 64.6%.
Which is better for knowledge tasks, Claude Opus 5 or GPT-5.6 Sol?
GPT-5.6 Sol has the edge for knowledge tasks in this comparison, averaging 94.6 versus 64.7. Inside this category, AA-Omniscience Hallucination Rate is the benchmark that creates the most daylight between them.
Which is better for coding, Claude Opus 5 or GPT-5.6 Sol?
Claude Opus 5 has the edge for coding in this comparison, averaging 89.5 versus 64.6. 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 Sol?
GPT-5.6 Sol has the edge for reasoning in this comparison, averaging 92.5 versus 90.4. Inside this category, ARC-AGI-3 is the benchmark that creates the most daylight between them.
Which is better for agentic tasks, Claude Opus 5 or GPT-5.6 Sol?
GPT-5.6 Sol has the edge for agentic tasks in this comparison, averaging 92 versus 90.8. Inside this category, AA Briefcase 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 Sol?
GPT-5.6 Sol has the edge for multimodal and grounded tasks in this comparison, averaging 83 versus 66.9. Inside this category, AA-MMMU-Pro is the benchmark that creates the most daylight between them.