Head-to-head comparison

Claude Opus 4.7 vs. Claude Opus 4.5

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

shared result series: 62
distinct benchmarks: 14
latest retrieval: 2026-10-15

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 Opus 4.7 and Claude Opus 4.5
Attribute	Claude Opus 4.7	Claude Opus 4.5
Provider	Anthropic	Anthropic
Released	Apr 2026	Nov 2025
Availability	Active	Active
Model type	Proprietary	Proprietary
Parameters	Not published	Not published
Architecture	Not published	Not published
Context window	1,000,000 tokens	200,000 tokens
Knowledge cutoff	January 2026	May 2025
Technical sources	Model, Knowledge cutoff	Model, Context

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 Opus 4.7 and Claude Opus 4.5
Price type	Claude Opus 4.7	Claude Opus 4.5
API input	$5	$5
API output	$25	$25
Cache read	$0.5	$0.5
Cache write (5 min.)	$6.25	$6.25
Cache write (1 hr.)	$10	$10
Price verified	09/22/2026 Source	09/22/2026 Source

Claude Opus 4.7Claude Opus 4.5

API inputUSD per 1M tokens, base tier

Claude Opus 4.7$5

Claude Opus 4.5$5

API outputUSD per 1M tokens, base tier

Claude Opus 4.7$25

Claude Opus 4.5$25

Cache readUSD per 1M tokens, base tier

Claude Opus 4.7$0.5

Claude Opus 4.5$0.5

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 Opus 4.7 and Claude Opus 4.5
Model	Provider	Purchase route and tariff	Input per 1M tokens	Output per 1M tokens	Source
Claude Opus 4.7	Anthropic	Direct from developerStandard API	Lowest listed price: $5	Lowest listed price: $25	Source09/22/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/eu-west-1	$5.5	$27.5	Source10/17/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/global	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Anthropic	Via OpenRouteranthropic	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Azure	Via OpenRouterazure/global	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Claude Platform on AWS	Via OpenRouterclaude-on-aws	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Google	Via OpenRoutergoogle-vertex/europe	$5.5	$27.5	Source10/17/2026
	Google	Via OpenRoutergoogle-vertex/global	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Google	Via OpenRoutergoogle-vertex/us	$5.5	$27.5	Source10/17/2026
Claude Opus 4.5	Anthropic	Direct from developerStandard API	Lowest listed price: $5	Lowest listed price: $25	Source09/22/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/eu-west-1	$5.5	$27.5	Source10/17/2026
	Anthropic	Via OpenRouteranthropic	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Azure	Via OpenRouterazure/global	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Claude Platform on AWS	Via OpenRouterclaude-on-aws	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026
	Google	Via OpenRoutergoogle-vertex/global	Lowest listed price: $5	Lowest listed price: $25	Source10/17/2026

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.

Not enough directly matched categories for a reliable radar

Only 1 of at least 5 categories have comparable, direct model measurements. The benchmark tables below show the available evidence without filling missing axes.

Claude Opus 4.7Claude Opus 4.5

CorpFin v2Overall. Higher is better. Test setup not documented as identical

Claude Opus 4.766.084 %

Claude Opus 4.561.305 %

GAIA ReliabilityAccuracy. Higher is better. No documented setup difference

Claude Opus 4.773.3 %

Claude Opus 4.568.5 %

GPQA DiamondOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.790.152 %

Claude Opus 4.579.546 %

LegalBenchOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.785.251 %

Claude Opus 4.582.837 %

LiveCodeBench v6Overall. Higher is better. Test setup not documented as identical

Claude Opus 4.785.073 %

Claude Opus 4.575.034 %

MedCodeOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.754.858 %

Claude Opus 4.545.174 %

MedScribeOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.782.953 %

Claude Opus 4.583.246 %

MMLU-ProOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.789.871 %

Claude Opus 4.585.59 %

MMMU-ProOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.785.549 %

Claude Opus 4.581.098 %

MortgageTaxOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.770.27 %

Claude Opus 4.568.68 %

SAGEOverall. Higher is better. Test setup not documented as identical

Claude Opus 4.756.103 %

Claude Opus 4.545.002 %

TaxEval v2Overall. Higher is better. Test setup not documented as identical

Claude Opus 4.775.266 %

Claude Opus 4.574.325 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Claude Opus 4.7 and Claude Opus 4.5
Benchmark and task	Claude Opus 4.7	Claude Opus 4.5	Source
CorpFin v2Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value66.084 %	61.305 %	Vals AI Source details Data as of: 08/12/2026
GAIA ReliabilityAccuracy Test details No documented difference in the test setupVersion: final leak-cleaned public validation setMetric: exact-match accuracyScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability165 GAIA tasks with repeated runs and perturbations. HAL does not publish cost for this evaluation.	Winner73.3 %	68.5 %	Princeton HAL Source details Data as of: 08/06/2026
GPQA DiamondOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value90.152 %	79.546 %	Vals AI Source details Data as of: 09/01/2026
LegalBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value85.251 %	82.837 %	Vals AI Source details Data as of: 10/13/2026
LiveCodeBench v6Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value85.073 %	75.034 %	Vals AI Source details Data as of: 09/01/2026
MedCodeOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 30,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 30,000 tokens	Higher measured value54.858 %	45.174 %	Vals AI Source details Data as of: 10/13/2026
MedScribeOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 30,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 30,000 tokens	82.953 %	Higher measured value83.246 %	Vals AI Source details Data as of: 10/13/2026
MMLU-ProOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value89.871 %	85.59 %	Vals AI Source details Data as of: 09/01/2026
MMMU-ProOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value85.549 %	81.098 %	Vals AI Source details Data as of: 09/01/2026
MortgageTaxOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value70.27 %	68.68 %	Vals AI Source details Data as of: 09/01/2026
SAGEOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 30,000 tokens	Higher measured value56.103 %	45.002 %	Vals AI Source details Data as of: 10/08/2026
TaxEval v2Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value75.266 %	74.325 %	Vals AI Source details Data as of: 09/01/2026
Terminal-Bench 2.0Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2.0Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: high, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value68.539 %	58.427 %	Vals AI Source details Data as of: 06/04/2026
τ-bench Airline CleanAccuracy Test details No documented difference in the test setupVersion: 26-task cleaned subsetMetric: task success rateScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability26 cleaned airline tasks. Princeton HAL recomputed every metric from scratch on this subset.	Winner84.6 %	80.8 %	Princeton HAL Source details Data as of: 08/05/2026
CorpFin v2Exact Pages Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value64.336 %	60.839 %	Vals AI Source details Data as of: 08/12/2026
GAIA ReliabilityConsistency Test details No documented difference in the test setupVersion: final leak-cleaned public validation setMetric: aggregate consistencyScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability165 GAIA tasks with repeated runs and perturbations. HAL does not publish cost for this evaluation.	Tie0.81 rating points	Tie0.81 rating points	Princeton HAL 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 Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value90.909 %	79.293 %	Vals AI Source details Data as of: 09/01/2026
LegalBenchConclusion Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value89.983 %	87.7 %	Vals AI Source details Data as of: 10/13/2026
LiveCodeBench v6Easy Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	No direct winner99.068 %	No direct winner99.068 %	Vals AI 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 Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value94.561 %	94.003 %	Vals AI Source details Data as of: 09/01/2026
MortgageTaxNumerical Extraction Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value78.696 %	77.424 %	Vals AI Source details Data as of: 09/01/2026
SAGEAutomata Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 30,000 tokens	Higher measured value43.65 %	29.365 %	Vals AI Source details Data as of: 10/08/2026
TaxEval v2Correctness Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value63.778 %	63.123 %	Vals AI Source details Data as of: 09/01/2026
Terminal-Bench 2.0Easy Test details Test setup not documented as identicalNo direct winnerVersion: 2.0Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: high, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	No direct winner100 %	No direct winner100 %	Vals AI Source details Data as of: 06/04/2026
τ-bench Airline CleanConsistency Test details No documented difference in the test setupVersion: 26-task cleaned subsetMetric: aggregate consistencyScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability26 cleaned airline tasks. Princeton HAL recomputed every metric from scratch on this subset.	Winner0.88 rating points	0.84 rating points	Princeton HAL Source details Data as of: 08/05/2026
CorpFin v2Max Fitting Context Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value67.832 %	60.723 %	Vals AI Source details Data as of: 08/12/2026
GAIA ReliabilityPredictability Test details No documented difference in the test setupVersion: final leak-cleaned public validation setMetric: aggregate predictabilityScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability165 GAIA tasks with repeated runs and perturbations. HAL does not publish cost for this evaluation.	0.73 rating points	Winner0.84 rating points	Princeton HAL 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 Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value89.394 %	79.798 %	Vals AI Source details Data as of: 09/01/2026
LegalBenchInterpretation Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value83.609 %	82.235 %	Vals AI Source details Data as of: 10/13/2026
LiveCodeBench v6Hard Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value66.857 %	44.571 %	Vals AI 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 Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value93.283 %	80.862 %	Vals AI Source details Data as of: 09/01/2026
MortgageTaxSemantic Extraction Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value61.844 %	59.936 %	Vals AI Source details Data as of: 09/01/2026
SAGECalculus Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Temperature: 1, Output limit: 30,000 tokens	Higher measured value75.926 %	58.333 %	Vals AI Source details Data as of: 10/08/2026
TaxEval v2Stepwise Reasoning Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value86.754 %	85.527 %	Vals AI Source details Data as of: 09/01/2026
Terminal-Bench 2.0Hard Test details Test setup not documented as identicalNo direct winnerVersion: 2.0Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: high, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value56.667 %	33.333 %	Vals AI Source details Data as of: 06/04/2026
τ-bench Airline CleanPredictability Test details No documented difference in the test setupVersion: 26-task cleaned subsetMetric: aggregate predictabilityScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability26 cleaned airline tasks. Princeton HAL recomputed every metric from scratch on this subset.	Winner0.87 rating points	0.85 rating points	Princeton HAL Source details Data as of: 08/05/2026
CorpFin v2Shared Max Context Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value66.084 %	62.354 %	Vals AI Source details Data as of: 08/12/2026
GAIA ReliabilityReliability Test details No documented difference in the test setupVersion: final leak-cleaned public validation setMetric: aggregate reliabilityScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability165 GAIA tasks with repeated runs and perturbations. HAL does not publish cost for this evaluation.	0.84 rating points	Winner0.85 rating points	Princeton HAL Source details Data as of: 08/06/2026
LegalBenchIssue Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value88.947 %	84.152 %	Vals AI Source details Data as of: 10/13/2026
LiveCodeBench v6Medium Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value89.295 %	81.462 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProChemistry Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value93.286 %	89.488 %	Vals AI Source details Data as of: 09/01/2026
SAGELinear Algebra Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 30,000 tokens	Higher measured value77.039 %	64.372 %	Vals AI Source details Data as of: 10/08/2026
Terminal-Bench 2.0Medium Test details Test setup not documented as identicalNo direct winnerVersion: 2.0Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: high, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value72.727 %	69.091 %	Vals AI Source details Data as of: 06/04/2026
τ-bench Airline CleanReliability Test details No documented difference in the test setupVersion: 26-task cleaned subsetMetric: aggregate reliabilityScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability26 cleaned airline tasks. Princeton HAL recomputed every metric from scratch on this subset.	Winner0.89 rating points	0.88 rating points	Princeton HAL Source details Data as of: 08/05/2026
GAIA ReliabilityRobustness Test details No documented difference in the test setupVersion: final leak-cleaned public validation setMetric: aggregate robustnessScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability165 GAIA tasks with repeated runs and perturbations. HAL does not publish cost for this evaluation.	Winner0.99 rating points	0.91 rating points	Princeton HAL Source details Data as of: 08/06/2026
LegalBenchRhetoric Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	78.722 %	Higher measured value79.777 %	Vals AI Source details Data as of: 10/13/2026
MMLU-ProComputer Science Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value92.683 %	84.146 %	Vals AI Source details Data as of: 09/01/2026
SAGEPartial Differential Equations Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 30,000 tokens	Higher measured value47.619 %	41.27 %	Vals AI Source details Data as of: 10/08/2026
τ-bench Airline CleanRobustness Test details No documented difference in the test setupVersion: 26-task cleaned subsetMetric: aggregate robustnessScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability26 cleaned airline tasks. Princeton HAL recomputed every metric from scratch on this subset.	0.91 rating points	Winner0.97 rating points	Princeton HAL Source details Data as of: 08/05/2026
GAIA ReliabilitySafety Test details No documented difference in the test setupVersion: final leak-cleaned public validation setMetric: aggregate safetyScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability165 GAIA tasks with repeated runs and perturbations. HAL does not publish cost for this evaluation.	Tie1 rating points	Tie1 rating points	Princeton HAL Source details Data as of: 08/06/2026
LegalBenchRule Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value84.996 %	80.32 %	Vals AI Source details Data as of: 10/13/2026
MMLU-ProEconomics Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value92.299 %	89.573 %	Vals AI Source details Data as of: 09/01/2026
τ-bench Airline CleanSafety Test details No documented difference in the test setupVersion: 26-task cleaned subsetMetric: aggregate safetyScoring: Higher is betterSettings: Agent: HAL Reliability, Harness: HAL Reliability26 cleaned airline tasks. Princeton HAL recomputed every metric from scratch on this subset.	0.94 rating points	Winner0.98 rating points	Princeton HAL Source details Data as of: 08/05/2026
MMLU-ProEngineering Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value88.029 %	76.161 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProHealth Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	82.396 %	Higher measured value83.985 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProHistory Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value83.202 %	78.478 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProLaw Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value82.107 %	78.292 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProMath Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value95.855 %	92.302 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProOthers Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value86.905 %	84.632 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProPhilosophy Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value88.978 %	87.776 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProPhysics Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value93.38 %	89.838 %	Vals AI Source details Data as of: 09/01/2026
MMLU-ProPsychology Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.7: Compute: max, Temperature: 1, Output limit: 128,000 tokens Claude Opus 4.5: Snapshot: claude-opus-4-5-20251101, Compute: high, Temperature: 1, Output limit: 64,000 tokens	Higher measured value91.228 %	88.722 %	Vals AI Source details Data as of: 09/01/2026
Key to the marks:better valuebetter, but not conclusivetie

Our model tests

Our tests

SVG short test

One SVG prompt and fixed output settings. Captures are shown only with 0 reported reasoning tokens or documented lack of reasoning controls. Identical test conditions do not make the models' internal compute identical.

Shared test round: 4ae5d8eeb9a9

Claude Opus 4.7

Unavailable in this round

Claude Opus 4.5

Unavailable in this round

Earlier tests

Pelican on a bicycle

A community classic for free-form SVG drawing.

Claude Opus 4.7

Reasoning (requested): high

Current attempt

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: anthropic/claude-opus-4.7
API model reported in response: anthropic/claude-opus-4.7
Reasoning setting: high
Pinned endpoint: Anthropic | anthropic/claude-4.7-opus-20260416
Temperature: Provider default
Token limit including reasoning: 8,192 tokens
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating
Response time: 19.9 s

Visually reviewed. This individual test does not establish overall model quality.

Claude Opus 4.5

Reasoning (requested): high

Current attempt

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: anthropic/claude-opus-4.5
API model reported in response: anthropic/claude-opus-4.5
Reasoning setting: high
Pinned endpoint: Anthropic | anthropic/claude-4.5-opus-20251124
Temperature: Provider default
Token limit including reasoning: 8,192 tokens
Test protocol: community-visual-tests-v1
Technical checks: Passed, not a quality rating
Response time: 21.6 s

Visually reviewed. This individual test does not establish overall model quality.

Prompt and test conditions

Generate an SVG of a pelican riding a bicycle

Token limit including reasoning: 8,192 tokens

Task origin (Simon Willison)

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 Opus 4.7 and Claude Opus 4.5 with other leading models using the same data and benchmark logic.

View the complete LLM comparison

Claude Opus 4.7 vs. Claude Opus 4.5

Change model selection

General specifications

API pricing compared

Shared benchmarks

Capability profile from matched benchmarks

Our tests

SVG short test

Claude Opus 4.7

Claude Opus 4.5

Pelican on a bicycle

Winning one benchmark is not an overall verdict

Related LLM comparisons