Head-to-head comparison

Claude Sonnet 5 vs. GPT-5.6 Sol

Compare published benchmark results from matching versions and cohorts, API pricing, and technical specifications. Any documented test setup differences remain visible.

shared result series: 200
distinct benchmarks: 26
latest retrieval: 2026-09-04

First modelSecond model

Specifications

General specifications

Release date, availability, context window, architecture, and published parameter counts at a glance. Undisclosed details remain marked as unavailable.

Specifications for Claude Sonnet 5 and GPT-5.6 Sol
Attribute	Claude Sonnet 5	GPT-5.6 Sol
Provider	Anthropic	OpenAI
Released	Jun 30, 2026	Jun 26, 2026
Availability	Active	Active
Model type	Proprietary	Proprietary
Parameters	Not published	Not published
Architecture	Not published	Not published
Context window	1,000,000 tokens	1,050,000 tokens
Knowledge cutoff	January 2026	February 16, 2026
Technical sources	Model, Context	Model, Context, Knowledge cutoff

Pricing

API pricing compared

The table compares official standard API rates per 1M tokens, including published cache pricing. Batch, fast, priority, regional surcharges, tool calls, and cloud platform pricing are excluded.

API pricing for Claude Sonnet 5 and GPT-5.6 Sol
Price type	Claude Sonnet 5	GPT-5.6 Sol
API input	$2	$4 ≤272K / $8 >272K
API output	$10	$20 ≤272K / $30 >272K
Cache read	$0.2	$0.4 ≤272K / $0.8 >272K
Cache write (5 min.)	$2.5	$5 ≤272K / $10 >272K
Cache write (1 hr.)	$4	Not listed
Price verified	09/02/2026 Source	08/24/2026 Source

Claude Sonnet 5GPT-5.6 Sol

API inputUSD per 1M tokens, base tier

Claude Sonnet 5$2

GPT-5.6 Sol$4

API outputUSD per 1M tokens, base tier

Claude Sonnet 5$10

GPT-5.6 Sol$20

Cache readUSD per 1M tokens, base tier

Claude Sonnet 5$0.2

GPT-5.6 Sol$0.4

Key to the marks:better valuebetter, but not conclusivetie

Pricing by provider

Direct developer pricing and provider endpoint tariffs routed through OpenRouter are listed separately. Prices are in USD per 1M tokens. Context-dependent direct tiers plus regional, Flex, Priority, and other OpenRouter tariffs remain individually identifiable.

Provider pricing for Claude Sonnet 5 and GPT-5.6 Sol
Model	Provider	Purchase route and tariff	Input per 1M tokens	Output per 1M tokens	Source
Claude Sonnet 5	Anthropic	Direct from developerStandard API	Lowest listed price: $2	Lowest listed price: $10	Source09/02/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/global	Lowest listed price: $2	Lowest listed price: $10	Source09/06/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-east-1	$2.2	$11	Source09/06/2026
	Anthropic	Via OpenRouteranthropic	Lowest listed price: $2	Lowest listed price: $10	Source09/06/2026
	Azure	Via OpenRouterazure/global	Lowest listed price: $2	Lowest listed price: $10	Source09/06/2026
	Azure	Via OpenRouterazure/us	Lowest listed price: $2	Lowest listed price: $10	Source09/06/2026
	Claude Platform on AWS	Via OpenRouterclaude-on-aws	Lowest listed price: $2	Lowest listed price: $10	Source09/06/2026
	Google	Via OpenRoutergoogle-vertex/europe	$2.2	$11	Source09/06/2026
	Google	Via OpenRoutergoogle-vertex/global	Lowest listed price: $2	Lowest listed price: $10	Source09/06/2026
	Google	Via OpenRoutergoogle-vertex/us	$2.2	$11	Source09/06/2026
GPT-5.6 Sol	OpenAI	Direct from developerUp to 272,000 context tokens	$4	$20	Source08/24/2026
	OpenAI	Direct from developerAbove 272,000 context tokens	$8	$30	Source08/24/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-east-1	$4.4	$22	Source09/06/2026
	Azure	Via OpenRouterazure	$5	$30	Source09/06/2026
	Azure	Via OpenRouterazure/eu	$5.5	$33	Source09/06/2026
	Azure	Via OpenRouterazure/us	$5.5	$33	Source09/06/2026
	OpenAI	Via OpenRouteropenai	$2	$10	Source09/06/2026
	OpenAI	Via OpenRouteropenai/fast	$4	$20	Source09/06/2026
	OpenAI	Via OpenRouteropenai/flex	Lowest listed price: $1	Lowest listed price: $5	Source09/06/2026

Performance

Shared benchmarks

Only published results with the same benchmark version, task, metric, and comparison cohort are paired. Different reasoning levels, agents, harnesses, or output limits appear directly in the table.

Capability profile from matched benchmarks

Each axis averages directly matched benchmark families. A value of 100 means the stronger value within this pair, not a universal quality score.

Claude Sonnet 5

GPT-5.6 Sol

Coding: 73.3 / 99.8 (5 benchmark families)

Reasoning: 93.4 / 100 (1 benchmark family)

Mathematics: 92.8 / 100 (1 benchmark family)

Knowledge: 98.3 / 100 (1 benchmark family)

Finance: 98.3 / 98.2 (5 benchmark families)

Legal: 94.5 / 83.3 (3 benchmark families)

Agentic tasks: 86.4 / 100 (2 benchmark families)

Multimodal: 93.4 / 100 (1 benchmark family)

27 matched rows from 25 benchmark families were considered. The radar shows 8 of 12 comparable categories.

Claude Sonnet 5GPT-5.6 Sol

Code MigrationOverall. Higher is better. No documented setup difference

Claude Sonnet 544.392 %

GPT-5.6 Sol52.916 %

CorpFin v2Overall. Higher is better. No documented setup difference

Claude Sonnet 566.977 %

GPT-5.6 Sol64.375 %

EMBOverall. Higher is better. No documented setup difference

Claude Sonnet 566.316 %

GPT-5.6 Sol72.339 %

Finance Agent (v2)Overall. Higher is better. No documented setup difference

Claude Sonnet 553.909 %

GPT-5.6 Sol53.756 %

GPQA DiamondOverall. Higher is better. No documented setup difference

Claude Sonnet 588.889 %

GPT-5.6 Sol95.202 %

Harvey's Legal Agent BenchmarkOverall · Task fully resolved. Higher is better. No documented setup difference

Claude Sonnet 55 %

GPT-5.6 Sol2.5 %

Legal Research BenchOverall · All-pass. Higher is better. No documented setup difference

Claude Sonnet 541.827 %

GPT-5.6 Sol48.077 %

LegalBenchOverall. Higher is better. No documented setup difference

Claude Sonnet 583.922 %

GPT-5.6 Sol86.965 %

LiveCodeBench v6Overall. Higher is better. No documented setup difference

Claude Sonnet 582.429 %

GPT-5.6 Sol82.604 %

LMArena Text, Style ControlOverall. Higher is better. Test setup not documented as identical

Claude Sonnet 51,462.16

GPT-5.6 Sol1,482.868

MedCodeOverall. Higher is better. No documented setup difference

Claude Sonnet 547.54 %

GPT-5.6 Sol43.974 %

MedScribeOverall. Higher is better. No documented setup difference

Claude Sonnet 576.054 %

GPT-5.6 Sol85.233 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Claude Sonnet 5 and GPT-5.6 Sol
Benchmark and task	Claude Sonnet 5	GPT-5.6 Sol	Source
Code MigrationOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	44.392 %	Winner52.916 %	Vals AI Source details Data as of: 09/03/2026
CorpFin v2Overall Test details No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better	Winner66.977 %	64.375 %	Vals AI Source details Data as of: 08/12/2026
EMBOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	66.316 %	Winner72.339 %	Vals AI Source details Data as of: 09/03/2026
Finance Agent (v2)Overall Test details No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better	Winner53.909 %	53.756 %	Vals AI Source details Data as of: 09/03/2026
GPQA DiamondOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	88.889 %	Winner95.202 %	Vals AI Source details Data as of: 09/01/2026
Harvey's Legal Agent BenchmarkOverall · Task fully resolved Test details No documented difference in the test setupVersion: 1Metric: task resolution rateScoring: Higher is better	Winner5 %	2.5 %	Vals AI Source details Data as of: 09/03/2026
Legal Research BenchOverall · All-pass Test details No documented difference in the test setupVersion: 1Metric: all-pass rateScoring: Higher is better	41.827 %	Winner48.077 %	Vals AI Source details Data as of: 09/03/2026
LegalBenchOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	83.922 %	Winner86.965 %	Vals AI Source details Data as of: 09/01/2026
LiveCodeBench v6Overall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	82.429 %	Winner82.604 %	Vals AI Source details Data as of: 09/01/2026
LMArena Text, Style ControlOverall Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Sonnet 5: Snapshot: claude-sonnet-5-high, Reasoning: high, Harness: Arena text, style control GPT-5.6 Sol: Snapshot: gpt-5.6-sol-xhigh, Reasoning: xhigh, Harness: Arena text, style controlClaude Sonnet 5: Style-controlled Arena rating from 31,096 anonymous pairwise votes.GPT-5.6 Sol: Style-controlled Arena rating from 23,153 anonymous pairwise votes.	1,462.16 95% confidence interval: ±4.745The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	Higher measured value1,482.868 95% confidence interval: ±5.12The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	LMArena Source details Data as of: 09/01/2026
MedCodeOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner47.54 %	43.974 %	Vals AI Source details Data as of: 09/03/2026
MedScribeOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	76.054 %	Winner85.233 %	Vals AI Source details Data as of: 09/03/2026
MMLU-ProOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	87.545 %	Winner89.1 %	Vals AI Source details Data as of: 09/01/2026
MMMU-ProOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	83.006 %	Winner88.844 %	Vals AI Source details Data as of: 09/01/2026
MortgageTaxOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	Winner70.032 %	67.289 %	Vals AI Source details Data as of: 09/01/2026
ProgramBenchOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	0 %	Winner1.5 %	Vals AI Source details Data as of: 09/01/2026
ProofBenchOverall Test details No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is better	77 %	Winner83 %	Vals AI Source details Data as of: 09/01/2026
Public Benefits Bench v1.1Overall Test details No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is better	66.035 %	Winner66.509 %	Vals AI Source details Data as of: 09/01/2026
SAGEOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	48.925 %	Winner52.563 %	Vals AI Source details Data as of: 09/03/2026
SkillsBenchOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is betterSettings: Harness: OpenHands	46.476 %	Winner54.1 %	Vals AI Source details Data as of: 09/01/2026
SWE-bench VerifiedOverall Test details No documented difference in the test setupVersion: 1Metric: accuracyScoring: Higher is better	79.6 %	Winner96.2 %	Vals AI Source details Data as of: 09/01/2026
TaxEval v2Overall Test details No documented difference in the test setupVersion: 2Metric: accuracyScoring: Higher is better	Winner75.634 %	74.775 %	Vals AI Source details Data as of: 09/01/2026
Terminal-Bench 2.1Overall Test details No documented difference in the test setupVersion: 2.1Metric: accuracyScoring: Higher is better	74.532 %	Winner85.768 %	Vals AI Source details Data as of: 09/03/2026
Vals IndexOverall Test details No documented difference in the test setupVersion: 2Metric: weighted index scoreScoring: Higher is better	59.61 %	Winner63.709 %	Vals AI Source details Data as of: 09/03/2026
Vals Multimodal IndexOverall Test details No documented difference in the test setupVersion: 1.2Metric: weighted index scoreScoring: Higher is better	68.826 %	Winner72.64 %	Vals AI Source details Data as of: 08/11/2026
Vibe Code Bench v1.1Overall Test details No documented difference in the test setupVersion: 1.1Metric: accuracyScoring: Higher is betterSettings: Harness: OpenHands	Winner81.325 %	80.495 %	Vals AI Source details Data as of: 09/03/2026
Code MigrationCLI Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	36.332 %	Higher measured value47.221 %	Vals AI Source details Data as of: 08/09/2026
CorpFin v2Exact Pages Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value63.869 %	61.538 %	Vals AI Source details Data as of: 08/06/2026
EMBComps Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	69.384 %	Higher measured value71.309 %	Vals AI Source details Data as of: 08/06/2026
Finance Agent (v2)Adjustments Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	48.715 %	Higher measured value49.233 %	Vals AI Source details Data as of: 08/06/2026
GPQA DiamondFew-Shot CoT Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	87.374 %	Higher measured value94.949 %	Vals AI Source details Data as of: 08/09/2026
Harvey's Legal Agent BenchmarkAntitrust Competition · Criteria passed Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: criteria pass rateScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value88.768 %	78.623 %	Vals AI Source details Data as of: 08/07/2026
Legal Research BenchAdministrative / Regulatory · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	54.545 %	Higher measured value60.606 %	Vals AI Source details Data as of: 08/06/2026
LegalBenchConclusion Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	89.555 %	Higher measured value90.492 %	Vals AI Source details Data as of: 08/09/2026
LiveCodeBench v6Easy Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value98.447 %	98.137 %	Vals AI Source details Data as of: 08/09/2026
LMArena Text, Style ControlBusiness, management and finance Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Sonnet 5: Snapshot: claude-sonnet-5-high, Reasoning: high, Harness: Arena text, style control GPT-5.6 Sol: Snapshot: gpt-5.6-sol-xhigh, Reasoning: xhigh, Harness: Arena text, style controlClaude Sonnet 5: Style-controlled Arena rating from 6,214 anonymous pairwise votes.GPT-5.6 Sol: Style-controlled Arena rating from 4,509 anonymous pairwise votes.	1,462.981 95% confidence interval: ±8.229The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	Higher measured value1,490.042 95% confidence interval: ±9.485The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	LMArena Source details Data as of: 09/01/2026
MMLU-ProBiology Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	93.584 %	Higher measured value94.561 %	Vals AI Source details Data as of: 08/09/2026
MortgageTaxNumerical Extraction Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value78.696 %	75.119 %	Vals AI Source details Data as of: 08/09/2026
ProgramBenchAlmost Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	13.5 %	Higher measured value23 %	Vals AI Source details Data as of: 08/09/2026
SAGEAutomata Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 30,000 tokens	34.92 %	Higher measured value39.682 %	Vals AI Source details Data as of: 08/09/2026
SkillsBenchNo Skills Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Harness: OpenHands, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens	35.123 %	Higher measured value43.312 %	Vals AI Source details Data as of: 08/07/2026
SWE-bench Verified<15 min fix Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	83.505 %	Higher measured value97.423 %	Vals AI Source details Data as of: 08/08/2026
TaxEval v2Correctness Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	64.023 %	Higher measured value65.004 %	Vals AI Source details Data as of: 08/09/2026
Terminal-Bench 2.1Easy Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value100 %	91.667 %	Vals AI Source details Data as of: 08/06/2026
Vals IndexCorpFin v2 Test details Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: benchmark scoreScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value67.949 %	66.434 %	Vals AI Source details Data as of: 08/09/2026
Vals Multimodal IndexCorpFin v2 Test details Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: benchmark scoreScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value67.949 %	66.434 %	Vals AI Source details Data as of: 08/09/2026
Code MigrationCOBOL Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	68.571 %	Higher measured value70 %	Vals AI Source details Data as of: 08/09/2026
CorpFin v2Max Fitting Context Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value69.114 %	65.152 %	Vals AI Source details Data as of: 08/06/2026
EMBDataroom Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	84.909 %	Higher measured value85.243 %	Vals AI Source details Data as of: 08/06/2026
Finance Agent (v2)All-Pass Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value42.084 %	41.362 %	Vals AI Source details Data as of: 08/06/2026
GPQA DiamondZero-Shot CoT Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	90.404 %	Higher measured value95.455 %	Vals AI Source details Data as of: 08/09/2026
Harvey's Legal Agent BenchmarkAntitrust Competition · Task fully resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: task resolution rateScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	No direct winner0 %	No direct winner0 %	Vals AI Source details Data as of: 08/07/2026
Legal Research BenchAdministrative / Regulatory · Weighted partial credit Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: weighted partial-credit scoreScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	87.81 %	Higher measured value90.69 %	Vals AI Source details Data as of: 08/06/2026
LegalBenchInterpretation Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	81.412 %	Higher measured value84.398 %	Vals AI Source details Data as of: 08/09/2026
LiveCodeBench v6Hard Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	66.857 %	Higher measured value67.429 %	Vals AI Source details Data as of: 08/09/2026
LMArena Text, Style ControlChinese Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Sonnet 5: Snapshot: claude-sonnet-5-high, Reasoning: high, Harness: Arena text, style control GPT-5.6 Sol: Snapshot: gpt-5.6-sol-xhigh, Reasoning: xhigh, Harness: Arena text, style controlClaude Sonnet 5: Style-controlled Arena rating from 2,074 anonymous pairwise votes.GPT-5.6 Sol: Style-controlled Arena rating from 1,557 anonymous pairwise votes.	1,499.492 95% confidence interval: ±13.64The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	Higher measured value1,534.658 95% confidence interval: ±15.764The Arena score measures human preference rather than a fixed capability test. The confidence interval and model configuration are part of the result.	LMArena Source details Data as of: 09/01/2026
MMLU-ProBusiness Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	91.635 %	Higher measured value93.283 %	Vals AI Source details Data as of: 08/09/2026
MortgageTaxSemantic Extraction Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value61.367 %	59.459 %	Vals AI Source details Data as of: 08/09/2026
ProgramBenchFully Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	0 %	Higher measured value1.5 %	Vals AI Source details Data as of: 08/09/2026
SAGELinear Algebra Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 30,000 tokens	68.999 %	Higher measured value70.388 %	Vals AI Source details Data as of: 08/09/2026
SkillsBenchWith Skills Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Harness: OpenHands, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens	46.476 %	Higher measured value54.1 %	Vals AI Source details Data as of: 08/07/2026
SWE-bench Verified>4 hours Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	66.667 %	Higher measured value100 %	Vals AI Source details Data as of: 08/08/2026
TaxEval v2Stepwise Reasoning Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	Higher measured value87.244 %	84.546 %	Vals AI Source details Data as of: 08/09/2026
Terminal-Bench 2.1Hard Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterClaude Sonnet 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Sol: Reasoning: max, Output limit: 128,000 tokens	57.778 %	Higher measured value84.444 %	Vals AI Source details Data as of: 08/06/2026
Key to the marks:better valuebetter, but not conclusivetie

How to read this comparison

Winning one benchmark is not an overall verdict

Your workload, budget, and required context length matter most. A coding benchmark says little about visual understanding or agent performance.

The standard price table uses direct developer rates. The provider comparison labels OpenRouter endpoints, regions, and special tariffs separately.

Every result links to its measurement source and retrieval date. Labels distinguish vendor reports, official benchmark leaderboards, and independent evaluations. Disputed or archived results are excluded.

More matchups

Compare Claude Sonnet 5 and GPT-5.6 Sol with other leading models using the same data and benchmark logic.

View the complete LLM comparison