# Claude Sonnet 5.5 Benchmark Scores & Performance

> Claude Sonnet 5.5 by Anthropic scores 80.49/100 overall, ranking #5 out of 512 AI models.

Canonical page: https://benchlm.ai/models/claude-sonnet-5-5

Last updated: September 28, 2026

## Model Details

| Property | Value |
|----------|-------|
| Creator | Anthropic |
| Source Type | Proprietary |
| Reasoning Type | Reasoning |
| Context Window | 1M |
| Overall Score | 80.49/100 |
| Overall Rank | #5 of 512 |

## Family & Coverage

- Family: Claude Sonnet 5.5
- Variant: base
- Benchmarks covered: 61 of 486
- Related earlier model: [Claude Sonnet 5](/models/claude-sonnet-5)
- Coverage note: BenchLM currently has partial benchmark coverage for this model, so the overall score is conservative.

## Agentic Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Terminal-Bench 4.0](/benchmarks/terminal-bench-4) | 70.60% |
| [GDPval-AA](/benchmarks/gdpvalaa) | 1844 |
| [HLE w/ tools](/benchmarks/hlewithtools) | 64.5% |
| [DRACO](/benchmarks/draco) | 87.0% |
| [Terminal-Bench-Science 0.1](/benchmarks/terminal-bench-science) | 59.9% |
| [LAB all-pass (Harvey held-out)](/benchmarks/legalagentbenchheldoutallpass) | 10.0% |
| [LAB criterion-pass (Harvey held-out)](/benchmarks/legalagentbenchheldoutcriterionpass) | 93.1% |
| [Toolathlon Verified Pass@3](/benchmarks/toolathlonverifiedpass3) | 85.2% |
| [Toolathlon Verified Pass³](/benchmarks/toolathlonverifiedpass3all) | 68.5% |
| [Toolathlon Verified avg. turns](/benchmarks/toolathlonverifiedavgturns) | 31.6 turns |
| [GDPval-AA](/benchmarks/gdpvalaanormalized) | 67.2% |
| [AA Briefcase](/benchmarks/aabriefcaseelo) | 1811 |
| [AA Harvey LAB](/benchmarks/aaharveylab) | 93.1% |
| [AA Terminal-Bench 4.0](/benchmarks/aaterminalbench4) | 63.6% |
| [GDP.pdf](/benchmarks/aagdppdf) | 25.8% |
| [AutomationBench (Zapier 1.0.6)](/benchmarks/automationbenchzapier106) | 44.7% |
| [Toolathlon-Verified](/benchmarks/toolathlonverified) | 77.8% |

## Coding Benchmarks

| Benchmark | Score |
|-----------|-------|
| [FrontierCode 1.1 Main](/benchmarks/frontiercode) | 46.2% |
| [CursorBench 4.0](/benchmarks/cursorbench40) | 55.5% |
| [SWE-bench Pro](/benchmarks/swe-bench-pro) | 81.3% |
| [SWE Multilingual](/benchmarks/swe-bench-multilingual) | 90.3% |
| [SWE Multimodal](/benchmarks/swe-bench-multimodal) | 54.3% |
| [DeepSWE](/benchmarks/deepswe) | 71.0% |
| [FrontierCode 1.1 Extended](/benchmarks/frontiercode11extended) | 59.1% |
| [ProgramBench](/benchmarks/programbench) | 79.7% |
| [FrontierSWE v2](/benchmarks/frontierswev2) | 61.9% |
| [AA-SciCode](/benchmarks/aascicode) | 61.0% |

## Multimodal & Grounded Benchmarks

| Benchmark | Score |
|-----------|-------|
| [Chartography (no tools)](/benchmarks/chartography) | 61.6% |
| [Chartography (tools)](/benchmarks/chartographywithtools) | 90.2% |
| [BenchCAD Vision2Code (no tools)](/benchmarks/benchcadvision2code) | 0.747 |
| [BenchCAD Vision2Code (tools)](/benchmarks/benchcadvision2codewithtools) | 0.963 |
| [Biomedical image analysis](/benchmarks/biomedicalimageanalysis) | 72.2% |
| [OfficeQA](/benchmarks/officeqa) | 76.9% |
| [OfficeQA Pro](/benchmarks/officeqapro) | 65.6% |

## Reasoning Benchmarks

| Benchmark | Score |
|-----------|-------|
| [AA-LCR](/benchmarks/lcr) | 82.7% |
| [CritPt](/benchmarks/critpt) | 31.4% |

## Knowledge Benchmarks

| Benchmark | Score |
|-----------|-------|
| [HealthBench (raw)](/benchmarks/healthbench) | 69.4% |
| [HealthBench (length-adjusted)](/benchmarks/healthbenchlengthadjusted) | 65.4% |
| [HealthBench Professional (raw)](/benchmarks/healthbenchprofessionalraw) | 77.1% |
| [HealthBench Professional](/benchmarks/healthbenchprofessional) | 69.2% |
| [BioMysteryBench (human-solvable)](/benchmarks/biomysterybenchhumansolvable) | 89.2% |
| [BioMysteryBench (human-difficult)](/benchmarks/biomysterybenchhumandifficult) | 44.7% |
| [SpatialBench Verified](/benchmarks/spatialbenchverified) | 72.5% |
| [SingleCellBench](/benchmarks/singlecellbench) | 59.1% |
| [Protein Design](/benchmarks/proteindesign) | 51.0% |
| [Morphology-to-molecule matching](/benchmarks/morphologytomolecule) | 25.0% |
| [Medicinal chemistry](/benchmarks/medicinalchemistry) | 65.3% |
| [Protein Design library ranking](/benchmarks/proteindesignlibraryranking) | 54.8% |
| [De novo protein-binder design](/benchmarks/proteinbinderdesign) | 82.3% |
| [Protocols (troubleshooting)](/benchmarks/protocolstroubleshooting) | 67.3% |
| [Protocols (understanding)](/benchmarks/protocolsunderstanding) | 66.6% |
| [HLE w/o tools](/benchmarks/hlenotools) | 56.9% |
| [Artificial Analysis Intelligence Index](/benchmarks/artificialanalysis) | 56.0% |
| [AA-HLE](/benchmarks/aahle) | 55.0% |
| [AA-Omniscience Index](/benchmarks/aaomniscienceindex) | 32.3% |
| [AA-Omniscience Accuracy](/benchmarks/omniscienceaccuracy) | 54.0% |
| [AA-Omniscience Hallucination Rate](/benchmarks/omnisciencehallucinationrate) | 47.0% |

## Multilingual Benchmarks

| Benchmark | Score |
|-----------|-------|
| [GMMLU](/benchmarks/gmmlu) | 92.1% |
| [MILU](/benchmarks/milu) | 91.6% |

## Mathematics Benchmarks

| Benchmark | Score |
|-----------|-------|
| [ArXivMath Aug. 2026 (no tools)](/benchmarks/arxivmathaugust2026) | 86.8% |
| [ArXivMath Aug. 2026 (tools)](/benchmarks/arxivmathaugust2026withtools) | 95.2% |

## Other Anthropic Models

- [Claude Opus 5.5](/models/claude-opus-5-5) - Score: 87.07
- [Claude Fable 5.1](/models/claude-fable-5-1) - Score: 82.96
- [Claude Opus 5](/models/claude-opus-5) - Score: 79.64
- [Claude Fable 5](/models/claude-fable) - Score: 78.89
- [Claude Opus 4.8](/models/claude-opus-4-8) - Score: 70.39
- [Claude Opus 4.7 (Adaptive)](/models/claude-opus-4-7-adaptive) - Score: 68.53
- [Claude Sonnet 5](/models/claude-sonnet-5) - Score: 66.98
- [Claude Opus 4.7](/models/claude-opus-4-7) - Score: 66.29
- [Claude Opus 4.6](/models/claude-opus-4-6) - Score: 63.61
- [Claude Sonnet 4.6](/models/claude-sonnet-4-6) - Score: 56.28
- [Claude Opus 4.5](/models/claude-opus-4-5) - Score: 55.03
- [Claude Opus 4.5 Thinking](/models/claude-opus-4-5-thinking) - Score: 54.95
- [Claude Sonnet 4.5](/models/claude-sonnet-4-5) - Score: 47.86
- [Claude Haiku 4.5](/models/claude-haiku-4-5) - Score: 42.48
- [Claude 4.1 Opus](/models/claude-4-1-opus) - Score: 38.42
- [Claude 4.1 Opus Thinking](/models/claude-4-1-opus-thinking) - Score: 36.56
- [Claude 4 Sonnet](/models/claude-4-sonnet) - Score: 36.05
- [Claude 3.5 Sonnet](/models/claude-3-5-sonnet) - Score: 29.99
- [Claude 3 Opus](/models/claude-3-opus) - Score: 28.67
- [Claude 3 Haiku](/models/claude-3-haiku) - Score: 14.8
- [Claude Mythos 5](/models/claude-mythos-5) - Score: not computed
- [Claude Opus 4.6 (Adaptive)](/models/claude-opus-4-6-thinking) - Score: not computed
- [Claude Mythos 5.1](/models/claude-mythos-5-1) - Score: not computed
- [Claude Mythos Preview](/models/claude-mythos-preview) - Score: not computed
- [Claude Haiku 4.5 Thinking](/models/claude-haiku-4-5-thinking) - Score: not computed
- [Claude Sonnet 4.5 Thinking](/models/claude-sonnet-4-5-thinking) - Score: not computed