Head-to-head comparison

Claude Opus 5 vs. GPT-6 Astra

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

shared result series: 301
distinct benchmark metrics: 41
latest retrieval: 2026-10-19

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 5 and GPT-6 Astra
Attribute	Claude Opus 5	GPT-6 Astra
Provider	Anthropic	OpenAI
Released	Jul 24, 2026	Sep 3, 2026
Availability	Active	Limited access
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	May 2026	April 30, 2026
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 5 and GPT-6 Astra
Price type	Claude Opus 5	GPT-6 Astra
API input	$5	$10 ≤272K / $20 >272K
API output	$25	$50 ≤272K / $75 >272K
Cache read	$0.5	$1 ≤272K / $2 >272K
Cache write (5 min.)	$6.25	$12.5 ≤272K / $25 >272K
Cache write (1 hr.)	$10	Not listed
Price verified	09/22/2026 Source	09/22/2026 Source

Claude Opus 5GPT-6 Astra

API inputUSD per 1M tokens, base tier

Claude Opus 5$5

GPT-6 Astra$10

API outputUSD per 1M tokens, base tier

Claude Opus 5$25

GPT-6 Astra$50

Cache readUSD per 1M tokens, base tier

Claude Opus 5$0.5

GPT-6 Astra$1

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 5 and GPT-6 Astra
Model	Provider	Purchase route and tariff	Input per 1M tokens	Output per 1M tokens	Source
Claude Opus 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	Source10/19/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/eu-west-1	$5.5	$27.5	Source10/19/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-east-1	$5.5	$27.5	Source10/19/2026
	Anthropic	Via OpenRouteranthropic	Lowest listed price: $5	Lowest listed price: $25	Source10/19/2026
	Anthropic	Via OpenRouteranthropic/fast	$10	$50	Source10/19/2026
	Azure	Via OpenRouterazure/global	Lowest listed price: $5	Lowest listed price: $25	Source10/19/2026
	Azure	Via OpenRouterazure/us	$5.5	$27.5	Source10/19/2026
	Claude Platform on AWS	Via OpenRouterclaude-on-aws	Lowest listed price: $5	Lowest listed price: $25	Source10/19/2026
	Google	Via OpenRoutergoogle-vertex/europe	$5.5	$27.5	Source10/19/2026
	Google	Via OpenRoutergoogle-vertex/global	Lowest listed price: $5	Lowest listed price: $25	Source10/19/2026
	Google	Via OpenRoutergoogle-vertex/us	$5.5	$27.5	Source10/19/2026
GPT-6 Astra	OpenAI	Direct from developerUp to 272,000 context tokens	$10	$50	Source09/22/2026
	OpenAI	Direct from developerAbove 272,000 context tokens	$20	$75	Source09/22/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-west-2	$11	$55	Source10/19/2026
	Azure	Via OpenRouterazure	$10	$50	Source10/19/2026
	Azure	Via OpenRouterazure/us	$11	$55	Source10/19/2026
	OpenAI	Via OpenRouteropenai	$10	$50	Source10/19/2026
	OpenAI	Via OpenRouteropenai/fast	$20	$100	Source10/19/2026
	OpenAI	Via OpenRouteropenai/flex	Lowest listed price: $5	Lowest listed price: $25	Source10/19/2026
	OpenAI	Via OpenRouteropenai/ultrafast	$60	$300	Source10/19/2026

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 Opus 5

GPT-6 Astra

Coding: 78.7 / 100 (4 benchmark families)

Reasoning: 90.1 / 99.5 (4 benchmark families)

Mathematics: 100 / 98.1 (1 benchmark family)

Knowledge: 91.9 / 99.6 (2 benchmark families)

Finance: 100 / 99.2 (1 benchmark family)

Legal: 100 / 98.9 (1 benchmark family)

Agentic tasks: 95.1 / 97 (4 benchmark families)

Multimodal: 96.1 / 100 (1 benchmark family)

96 matched rows from 14 benchmark families were considered. The radar shows 8 of 11 comparable categories.

Claude Opus 5GPT-6 Astra

AA-AnalystAgentOverall. Higher is better. No documented setup difference

Claude Opus 553.8%

GPT-6 Astra51.2%

AA-BriefcaseOverall. Higher is better. No documented setup difference

Claude Opus 51,633.85 elo

GPT-6 Astra1,546.7 elo

AA-LCROverall. Higher is better. No documented setup difference

Claude Opus 580.3%

GPT-6 Astra80.0%

AA-OmniscienceOverall. Higher is better. No documented setup difference

Claude Opus 535.383 points

GPT-6 Astra43.417 points

AA-Omniscience AccuracyOverall. Higher is better. No documented setup difference

Claude Opus 559.5%

GPT-6 Astra61.9%

AA-Omniscience Hallucination RateOverall. Lower is better. No documented setup difference

Claude Opus 559.5%

GPT-6 Astra48.3%

Artificial Analysis Intelligence IndexOverall. Higher is better. No documented setup difference

Claude Opus 549.677 points

GPT-6 Astra52.386 points

AutomationBench-AAOverall. Higher is better. No documented setup difference

Claude Opus 553.2%

GPT-6 Astra67.2%

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

Claude Opus 579.259 %

GPT-6 Astra79.259 %

Code MigrationOverall. Higher is better. Test setup not documented as identical

Claude Opus 557.473 %

GPT-6 Astra67.74 %

CyberBench v1.1Overall. Higher is better. Test setup not documented as identical

Claude Opus 565.357 %

GPT-6 Astra41.072 %

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

Claude Opus 573.556 %

GPT-6 Astra71.7 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Claude Opus 5 and GPT-6 Astra
Benchmark and task	Claude Opus 5	GPT-6 Astra	Source
AA-AnalystAgentOverall Test details No documented difference in the test setupVersion: AA cohort: 80 questionsMetric: pass^5 reliabilityScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Max), Reasoning: Max, Agent: Stirrup analyst agent, Harness: Artificial Analysis Stirrup, Attempts: pass^5, 5 independent attempts per question GPT-6 Astra: Snapshot: GPT-6 Astra (Max), Reasoning: Max, Agent: Stirrup analyst agent, Harness: Artificial Analysis Stirrup, Attempts: pass^5, 5 independent attempts per questionClaude Opus 5: Exact AA configuration: Claude Opus 5 (Max) (release claude-opus-5, row claude-opus-5, UUID b8fc61f7-5e9a-49e6-8547-6ac56db24627, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption. 80 quantitative questions across 14 domains are run five independent times. pass^5 counts only questions answered correctly in all five attempts.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Max) (release gpt-6-astra, row gpt-6-astra, UUID 2f339a97-9a0d-499a-9cb5-e0db665bfa25, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption. 80 quantitative questions across 14 domains are run five independent times. pass^5 counts only questions answered correctly in all five attempts.	Winner53.8% Questions, source files, and reference answers are private. pass^5 measures reliability across all five attempts, not pass@1.	51.2% Questions, source files, and reference answers are private. pass^5 measures reliability across all five attempts, not pass@1.	Artificial Analysis Source details Data as of: 10/19/2026
AA-BriefcaseOverall Test details No documented difference in the test setupVersion: v1.1Metric: AA-Briefcase EloScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	Winner1,633.85 elo	1,546.7 elo	Artificial Analysis Source details Data as of: 10/19/2026
AA-LCROverall Test details No documented difference in the test setupVersion: v1.1Metric: Pass rateScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	Winner80.3%	80.0%	Artificial Analysis Source details Data as of: 10/19/2026
AA-OmniscienceOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: AA-Omniscience IndexScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	35.383 points The index can be negative and ranges from -100 to 100 according to Artificial Analysis.	Winner43.417 points The index can be negative and ranges from -100 to 100 according to Artificial Analysis.	Artificial Analysis Source details Data as of: 10/19/2026
AA-Omniscience AccuracyOverall Test details No documented difference in the test setupVersion: AA Omniscience diagnosticMetric: AA-Omniscience AccuracyScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: Xhigh, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboardClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption. Proportion of all questions answered correctly, regardless of whether the model answers or abstains on other questions.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption. Proportion of all questions answered correctly, regardless of whether the model answers or abstains on other questions.	59.5% This is the separately published accuracy diagnostic, not the AA-Omniscience Index. AA identifies higher values as better.	Winner61.9% This is the separately published accuracy diagnostic, not the AA-Omniscience Index. AA identifies higher values as better.	Artificial Analysis Source details Data as of: 10/19/2026
AA-Omniscience Hallucination RateOverall Test details No documented difference in the test setupVersion: AA Omniscience diagnosticMetric: AA-Omniscience Hallucination RateScoring: Lower is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: Xhigh, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboardClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).	59.5% This is the separately published hallucination-rate diagnostic, not the AA-Omniscience Index. AA identifies lower values as better.	Winner48.3% This is the separately published hallucination-rate diagnostic, not the AA-Omniscience Index. AA identifies lower values as better.	Artificial Analysis Source details Data as of: 10/19/2026
Artificial Analysis Intelligence IndexOverall Test details No documented difference in the test setupVersion: v4.3.2Metric: Artificial Analysis Intelligence IndexScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	49.677 points	Winner52.386 points	Artificial Analysis Source details Data as of: 10/19/2026
AutomationBench-AAOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: Objectives completed without guardrail violationsScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	53.2%	Winner67.2%	Artificial Analysis Source details Data as of: 10/19/2026
BioMysteryBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Harness: Terminus 2, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Harness: Terminus 2, Output limit: 128,000 tokens	No direct winner79.259 %	No direct winner79.259 %	Vals AI Source details Data as of: 10/12/2026
Code MigrationOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	57.473 %	Higher measured value67.74 %	Vals AI Source details Data as of: 10/12/2026
CyberBench v1.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value65.357 %	41.072 %	Vals AI Source details Data as of: 10/12/2026
EMBOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value73.556 %	71.7 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value58.633 %	53.54 %	Vals AI Source details Data as of: 10/12/2026
GDP.pdfOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: All-pass rateScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	21.0%	Winner32.2%	Artificial Analysis Source details Data as of: 10/19/2026
GDPval-AAOverall Test details No documented difference in the test setupVersion: v2.1Metric: GDPval-AA EloScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	Winner1,690.67 elo	1,555.91 elo	Artificial Analysis Source details Data as of: 10/19/2026
Harvey's Legal Agent BenchmarkOverall · Task fully resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: task resolution rateScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value6.667 %	5.417 %	Vals AI Source details Data as of: 10/12/2026
Humanity's Last ExamOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: Text-only pass@1 accuracyScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	54.4% Artificial Analysis describes this comparable run as 2,158 text-only questions, not the full 2,500-question dataset.	Winner54.6% Artificial Analysis describes this comparable run as 2,158 text-only questions, not the full 2,500-question dataset.	Artificial Analysis Source details Data as of: 10/19/2026
IOIOverall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	84.333 %	Higher measured value100 %	Vals AI Source details Data as of: 10/14/2026
Legal Research BenchOverall · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value55.288 %	39.423 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevOverall Test details Test setup not documented as identicalNo direct winnerVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 5: Snapshot: claude-opus-5-high, Reasoning: high, Harness: Arena Code Arena WebDev GPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 5: Arena rating from 23,519 published pairwise votes.GPT-6 Astra: Arena rating from 5,746 published pairwise votes.	1,657.837 95% confidence interval: ±5.533The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	Higher measured value1,780.244 95% confidence interval: ±10.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena SearchOverall Test details Test setup not documented as identicalNo direct winnerVersion: searchMetric: Search Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 5: Snapshot: claude-opus-5-high, Reasoning: high, Harness: Arena Search GPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena SearchClaude Opus 5: Arena rating from 13,285 published pairwise votes.GPT-6 Astra: Arena rating from 18,483 published pairwise votes.	Higher measured value1,241.568 95% confidence interval: ±6.871The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,233.616 95% confidence interval: ±7.842The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
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 Opus 5: Snapshot: claude-opus-5-high, Reasoning: high, Harness: Arena Text, style control GPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style controlClaude Opus 5: Arena rating from 65,663 published pairwise votes.GPT-6 Astra: Arena rating from 11,674 published pairwise votes.	Higher measured value1,489.847 95% confidence interval: ±3.689The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,476.716 95% confidence interval: ±6.385The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlOverall Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 5: Snapshot: claude-opus-5-high, Reasoning: high, Harness: Arena Vision, style control GPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style controlClaude Opus 5: Arena rating from 16,665 published pairwise votes.GPT-6 Astra: Arena rating from 3,590 published pairwise votes.	Higher measured value1,287.417 95% confidence interval: ±6.783The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,285.988 95% confidence interval: ±10.782The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
MedCodeOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value63.57 %	48.486 %	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 5: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value90.985 %	87.908 %	Vals AI Source details Data as of: 10/13/2026
MMMU-ProOverall Test details No documented difference in the test setupVersion: AA cohort: 1,730 questionsMetric: pass@1 accuracyScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh, Attempts: pass@1, one run per question GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: Xhigh, Attempts: pass@1, one run per questionClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption. The AA run covers 1,730 questions with ten answer options. Answers are extracted by regex and scored pass@1.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption. The AA run covers 1,730 questions with ten answer options. Answers are extracted by regex and scored pass@1.	84.0% The benchmark page describes 3,460 questions in the full MMMU-Pro dataset; AA's methodology specifies a 1,730-question run.	Winner86.2% The benchmark page describes 3,460 questions in the full MMMU-Pro dataset; AA's methodology specifies a 1,730-question run.	Artificial Analysis Source details Data as of: 10/19/2026
MysteryMechanismOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	37.387 %	Higher measured value53.153 %	Vals AI Source details Data as of: 10/12/2026
ProgramBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: fully resolved rateScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	3 %	Higher measured value5.5 %	Vals AI Source details Data as of: 10/12/2026
ProofBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	No direct winner99 %	No direct winner99 %	Vals AI Source details Data as of: 10/12/2026
SAGEOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 30,000 tokens	Higher measured value49.426 %	46.367 %	Vals AI Source details Data as of: 10/08/2026
SRE BenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: fully solved rateScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	12.214 %	Higher measured value56.87 %	Vals AI Source details Data as of: 10/14/2026
Tax Agent BenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value45.548 %	20.717 %	Vals AI Source details Data as of: 10/12/2026
Terminal-BenchOverall Test details No documented difference in the test setupVersion: 4.0Metric: Task successScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Xhigh), Reasoning: Xhigh GPT-6 Astra: Snapshot: GPT-6 Astra (Xhigh), Reasoning: XhighClaude Opus 5: Exact AA configuration: Claude Opus 5 (Xhigh) (release claude-opus-5, row claude-opus-5-xhigh, UUID 1305c921-7aaa-4d6d-99b5-99b3acf15e19, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Xhigh) (release gpt-6-astra, row gpt-6-astra-xhigh, UUID 1f541ef3-913f-4eb2-9d07-0e93c7a9a5e3, reasoning=true, source reasoning-token configuration budget=0, suffix=Xhigh). This configured value is not measured token consumption.	46.5%	Winner59.6%	Artificial Analysis Source details Data as of: 10/19/2026
Terminal-Bench 2.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: high, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	84.644 %	Higher measured value87.266 %	Vals AI Source details Data as of: 09/28/2026
Terminal-Bench 4.0Overall Test details Test setup not documented as identicalNo direct winnerVersion: 4.0Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	53.535 %	Higher measured value59.596 %	Vals AI Source details Data as of: 10/13/2026
Terminal-Bench ScienceOverall Test details Test setup not documented as identicalNo direct winnerVersion: 0.1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Harness: Mini-SWE-agent, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Harness: Mini-SWE-agent, Output limit: 128,000 tokens	27.143 %	Higher measured value62.857 %	Vals AI Source details Data as of: 10/14/2026
Terminal-Bench-ScienceOverall Test details No documented difference in the test setupVersion: 0.1.0Metric: pass@1 task success rateScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Max), Reasoning: Max, Agent: mini-swe-agent, Harness: mini-swe-agent, Attempts: pass@1, averaged over 3 repeats per task GPT-6 Astra: Snapshot: GPT-6 Astra (Max), Reasoning: Max, Agent: mini-swe-agent, Harness: mini-swe-agent, Attempts: pass@1, averaged over 3 repeats per taskClaude Opus 5: Exact AA configuration: Claude Opus 5 (Max) (release claude-opus-5, row claude-opus-5, UUID b8fc61f7-5e9a-49e6-8547-6ac56db24627, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption. The full 0.1.0 release has 70 tasks across five scientific domains. A task passes only when its entire test suite passes. AA averages pass@1 over three repeats.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Max) (release gpt-6-astra, row gpt-6-astra, UUID 2f339a97-9a0d-499a-9cb5-e0db665bfa25, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption. The full 0.1.0 release has 70 tasks across five scientific domains. A task passes only when its entire test suite passes. AA averages pass@1 over three repeats.	28.6%	Winner63.3%	Artificial Analysis Source details Data as of: 10/19/2026
Time Horizon Index: KSPOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: progressScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	18.833 %	Higher measured value90.5 %	Vals AI Source details Data as of: 10/07/2026
Vals IndexOverall Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: weighted index scoreScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value63.674 %	63.125 %	Vals AI Source details Data as of: 10/13/2026
Vibe Code Bench 1-100Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.0Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Harness: OpenHands, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens	Higher measured value28.533 %	27.644 %	Vals AI Source details Data as of: 10/14/2026
Vibe Code Bench v1.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Opus 5: Harness: OpenHands GPT-6 Astra: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens	88.404 %	Higher measured value89.594 %	Vals AI Source details Data as of: 10/12/2026
AA-BriefcaseOverall Test details No documented difference in the test setupVersion: v1.1Metric: AA-Briefcase EloScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	1,209.28 elo	Winner1,257.01 elo	Artificial Analysis Source details Data as of: 10/19/2026
AA-LCROverall Test details No documented difference in the test setupVersion: v1.1Metric: Pass rateScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	Winner81.3%	80.0%	Artificial Analysis Source details Data as of: 10/19/2026
AA-OmniscienceOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: AA-Omniscience IndexScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	28.55 points The index can be negative and ranges from -100 to 100 according to Artificial Analysis.	Winner40.55 points The index can be negative and ranges from -100 to 100 according to Artificial Analysis.	Artificial Analysis Source details Data as of: 10/19/2026
AA-Omniscience AccuracyOverall Test details No documented difference in the test setupVersion: AA Omniscience diagnosticMetric: AA-Omniscience AccuracyScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: Low, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboardClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption. Proportion of all questions answered correctly, regardless of whether the model answers or abstains on other questions.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption. Proportion of all questions answered correctly, regardless of whether the model answers or abstains on other questions.	56.0% This is the separately published accuracy diagnostic, not the AA-Omniscience Index. AA identifies higher values as better.	Winner59.5% This is the separately published accuracy diagnostic, not the AA-Omniscience Index. AA identifies higher values as better.	Artificial Analysis Source details Data as of: 10/19/2026
AA-Omniscience Hallucination RateOverall Test details No documented difference in the test setupVersion: AA Omniscience diagnosticMetric: AA-Omniscience Hallucination RateScoring: Lower is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: Low, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboardClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).	62.2% This is the separately published hallucination-rate diagnostic, not the AA-Omniscience Index. AA identifies lower values as better.	Winner46.9% This is the separately published hallucination-rate diagnostic, not the AA-Omniscience Index. AA identifies lower values as better.	Artificial Analysis Source details Data as of: 10/19/2026
Artificial Analysis Intelligence IndexOverall Test details No documented difference in the test setupVersion: v4.3.2Metric: Artificial Analysis Intelligence IndexScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	39.354 points	Winner45.782 points	Artificial Analysis Source details Data as of: 10/19/2026
AutomationBench-AAOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: Objectives completed without guardrail violationsScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	51.8%	Winner59.1%	Artificial Analysis Source details Data as of: 10/19/2026
BioMysteryBenchHard Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Harness: Terminus 2, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Harness: Terminus 2, Output limit: 128,000 tokens	Higher measured value43.137 %	37.255 %	Vals AI Source details Data as of: 10/12/2026
Code MigrationCLI Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	53.297 %	Higher measured value67.463 %	Vals AI Source details Data as of: 10/12/2026
CyberBench v1.1Patch Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value85.714 %	82.143 %	Vals AI Source details Data as of: 10/12/2026
EMBComps Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	75.171 %	Higher measured value77.953 %	Vals AI Source details Data as of: 10/12/2026
Finance Agent (v2)Adjustments Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value54.617 %	48.318 %	Vals AI Source details Data as of: 10/12/2026
GDP.pdfOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: All-pass rateScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	17.2%	Winner30.4%	Artificial Analysis Source details Data as of: 10/19/2026
GDPval-AAOverall Test details No documented difference in the test setupVersion: v2.1Metric: GDPval-AA EloScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	1,302.46 elo	Winner1,389.25 elo	Artificial Analysis Source details Data as of: 10/19/2026
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 Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	87.319 %	Higher measured value88.768 %	Vals AI Source details Data as of: 10/12/2026
Humanity's Last ExamOverall Test details No documented difference in the test setupVersion: unspecifiedMetric: Text-only pass@1 accuracyScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: LowClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption.	43.4% Artificial Analysis describes this comparable run as 2,158 text-only questions, not the full 2,500-question dataset.	Winner49.2% Artificial Analysis describes this comparable run as 2,158 text-only questions, not the full 2,500-question dataset.	Artificial Analysis Source details Data as of: 10/19/2026
IOIIOI 2024 Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	No direct winner100 %	No direct winner100 %	Vals AI Source details Data as of: 10/14/2026
Legal Research BenchAdministrative / Regulatory · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	Higher measured value65.152 %	53.03 %	Vals AI Source details Data as of: 10/12/2026
LMArena Code Arena WebDevOverall Test details No documented difference in the test setupVersion: webdevMetric: WebDev Arena Bradley-Terry ratingScoring: Higher is betterClaude Opus 5: Snapshot: claude-opus-5-max, Reasoning: max, Harness: Arena Code Arena WebDev GPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Code Arena WebDevClaude Opus 5: Arena rating from 17,934 published pairwise votes.GPT-6 Astra: Arena rating from 5,746 published pairwise votes.	1,689.967 95% confidence interval: ±6.28The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	Winner1,780.244 95% confidence interval: ±10.541The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Text, Style ControlOverall Test details No documented difference in the test setupVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 5: Snapshot: claude-opus-5-max, Reasoning: max, Harness: Arena Text, style control GPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Text, style controlClaude Opus 5: Arena rating from 31,907 published pairwise votes.GPT-6 Astra: Arena rating from 11,674 published pairwise votes.	Winner1,489.524 95% confidence interval: ±4.846The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,476.716 95% confidence interval: ±6.385The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/13/2026
LMArena Vision, Style ControlChinese Test details Test setup not documented as identicalNo direct winnerVersion: style-control-v1Metric: style-controlled Bradley-Terry ratingScoring: Higher is betterClaude Opus 5: Snapshot: claude-opus-5-high, Reasoning: high, Harness: Arena Vision, style control GPT-6 Astra: Snapshot: gpt-6-astra-max, Reasoning: max, Harness: Arena Vision, style controlClaude Opus 5: Arena rating from 985 published pairwise votes.GPT-6 Astra: Arena rating from 230 published pairwise votes.	Higher measured value1,366.609 95% confidence interval: ±20.761The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,357.126 95% confidence interval: ±39.572The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	LMArena Source details Data as of: 10/09/2026
MMMU-ProOverall Test details No documented difference in the test setupVersion: AA cohort: 1,730 questionsMetric: pass@1 accuracyScoring: Higher is betterClaude Opus 5: Snapshot: Claude Opus 5 (Low), Reasoning: Low, Attempts: pass@1, one run per question GPT-6 Astra: Snapshot: GPT-6 Astra (Low), Reasoning: Low, Attempts: pass@1, one run per questionClaude Opus 5: Exact AA configuration: Claude Opus 5 (Low) (release claude-opus-5, row claude-opus-5-low, UUID 20928ba9-3a3f-415f-9519-b84ff64ecf34, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption. The AA run covers 1,730 questions with ten answer options. Answers are extracted by regex and scored pass@1.GPT-6 Astra: Exact AA configuration: GPT-6 Astra (Low) (release gpt-6-astra, row gpt-6-astra-low, UUID a3f8100d-e38f-408b-b0fa-0085dae18dc1, reasoning=true, source reasoning-token configuration budget=0, suffix=Low). This configured value is not measured token consumption. The AA run covers 1,730 questions with ten answer options. Answers are extracted by regex and scored pass@1.	79.8% The benchmark page describes 3,460 questions in the full MMMU-Pro dataset; AA's methodology specifies a 1,730-question run.	Winner84.6% The benchmark page describes 3,460 questions in the full MMMU-Pro dataset; AA's methodology specifies a 1,730-question run.	Artificial Analysis Source details Data as of: 10/19/2026
ProgramBenchAlmost Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: almost-resolved rateScoring: Higher is betterClaude Opus 5: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-6 Astra: Reasoning: max, Output limit: 128,000 tokens	41.5 %	Higher measured value50 %	Vals AI Source details Data as of: 10/12/2026
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 5

Unavailable in this round

GPT-6 Astra

Unavailable in this round

Earlier tests

Pelican on a bicycle

A community classic for free-form SVG drawing.

Claude Opus 5

Reasoning (requested): high

Current attempt

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: anthropic/claude-opus-5
API model reported in response: anthropic/claude-opus-5
Reasoning setting: high
Pinned endpoint: Anthropic | anthropic/claude-opus-5-20260723
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: 63 s

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

Earlier attempts (1)

Sep 7, 2026API request failedReasoning setting: high

GPT-6 Astra

Reasoning (requested): high

Current attempt

Run date: Sep 7, 2026
API provider: openrouter
Requested API model: openai/gpt-6-astra
API model reported in response: openai/gpt-6-astra
Reasoning setting: high
Pinned endpoint: OpenAI | openai/gpt-6-astra-20260903
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: 141.7 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 5 and GPT-6 Astra with other leading models using the same data and benchmark logic.

View the complete LLM comparison

Claude Opus 5 vs. GPT-6 Astra

Change model selection

General specifications

API pricing compared

Shared benchmarks

Capability profile from matched benchmarks

Our tests

SVG short test

Claude Opus 5

GPT-6 Astra

Pelican on a bicycle

Winning one benchmark is not an overall verdict

Related LLM comparisons