Head-to-head comparison

Claude Opus 4.8 vs. GPT-5.6 Luna

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

shared result series: 283
distinct benchmark metrics: 40
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 4.8 and GPT-5.6 Luna
Attribute	Claude Opus 4.8	GPT-5.6 Luna
Provider	Anthropic	OpenAI
Released	May 2026	Jul 9, 2026
Availability	Legacy	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, Knowledge cutoff	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 Opus 4.8 and GPT-5.6 Luna
Price type	Claude Opus 4.8	GPT-5.6 Luna
API input	$5	$0.2 ≤272K / $0.4 >272K
API output	$25	$1.2 ≤272K / $1.8 >272K
Cache read	$0.5	$0.02 ≤272K / $0.04 >272K
Cache write (5 min.)	$6.25	$0.25 ≤272K / $0.5 >272K
Cache write (1 hr.)	$10	Not listed
Price verified	09/22/2026 Source	09/22/2026 Source

Claude Opus 4.8GPT-5.6 Luna

API inputUSD per 1M tokens, base tier

Claude Opus 4.8$5

GPT-5.6 Luna$0.2

API outputUSD per 1M tokens, base tier

Claude Opus 4.8$25

GPT-5.6 Luna$1.2

Cache readUSD per 1M tokens, base tier

Claude Opus 4.8$0.5

GPT-5.6 Luna$0.02

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.

No provider tariff for Claude Opus 4.8 is available in the stored OpenRouter snapshot. The official direct price is still shown.

Provider pricing for Claude Opus 4.8 and GPT-5.6 Luna
Model	Provider	Purchase route and tariff	Input per 1M tokens	Output per 1M tokens	Source
Claude Opus 4.8	Anthropic	Direct from developerStandard API	Lowest listed price: $5	Lowest listed price: $25	Source09/22/2026
GPT-5.6 Luna	OpenAI	Direct from developerUp to 272,000 context tokens	$0.2	$1.2	Source09/22/2026
	OpenAI	Direct from developerAbove 272,000 context tokens	$0.4	$1.8	Source09/22/2026
	Amazon Bedrock	Via OpenRouteramazon-bedrock/us-east-1	$0.22	$1.32	Source10/26/2026
	Azure	Via OpenRouterazure	$0.2	$1.2	Source10/26/2026
	Azure	Via OpenRouterazure/eu	$0.22	$1.32	Source10/26/2026
	Azure	Via OpenRouterazure/us	$0.22	$1.32	Source10/26/2026
	OpenAI	Via OpenRouteropenai	$0.2	$1.2	Source10/26/2026
	OpenAI	Via OpenRouteropenai/fast	$0.4	$2.4	Source10/26/2026
	OpenAI	Via OpenRouteropenai/flex	Lowest listed price: $0.1	Lowest listed price: $0.6	Source10/26/2026

Benchmark index

Benchmark index compared

The index combines independent benchmarks on a scale from 0 to 100. A lead only counts when a model comes first in at least 95% of 200 recalculations with randomly redrawn test families. Otherwise both count as level.

Open the AI leaderboard

Claude Opus 4.8

Overall index

57.0

GPT-5.6 Luna

Overall index

36.2

Claude Opus 4.8 leads in 8 of 11 shared areas. 3 areas are tied.

Area	Claude Opus 4.8	GPT-5.6 Luna
Codingtied	55.3	43.1
Reasoningtied	46.2	51.1
Knowledge	65.6	15.3
Agents	56.7	47.6
Generaltied	45.3	30.3
Finance	65.0	52.8
Legal	57.4	29.8
Health	67.8	34.5
Multimodal	55.0	28.3
Productivity	68.3	35.7
Education	71.1	37.4

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 4.8

GPT-5.6 Luna

Coding: 100 / 53.5 (1 benchmark family)

Reasoning: 95.9 / 93.7 (3 benchmark families)

Knowledge: 100 / 43.3 (1 benchmark family)

Agentic tasks: 96.3 / 100 (3 benchmark families)

Search: 100 / 98.8 (1 benchmark family)

Overall: 97.9 / 94.7 (2 benchmark families)

13 matched rows from 11 benchmark families were considered. The radar shows 6 of 6 comparable categories.

Claude Opus 4.8GPT-5.6 Luna

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

Claude Opus 4.81,321.43 elo

GPT-5.6 Luna1,342.18 elo

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

Claude Opus 4.877.7%

GPT-5.6 Luna83.7%

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

Claude Opus 4.828.75 points

GPT-5.6 Luna-10.283 points

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

Claude Opus 4.848.8%

GPT-5.6 Luna42.7%

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

Claude Opus 4.839.3%

GPT-5.6 Luna92.6%

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

Claude Opus 4.841.79 points

GPT-5.6 Luna37.324 points

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

Claude Opus 4.845.6%

GPT-5.6 Luna50.2%

BrowseCompOverall. Higher is better. No documented setup difference

Claude Opus 4.884.3 %

GPT-5.6 Luna83.3 %

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

Claude Opus 4.847.25 %

GPT-5.6 Luna44.55 %

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

Claude Opus 4.866.706 %

GPT-5.6 Luna64.219 %

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

Claude Opus 4.869.373 %

GPT-5.6 Luna67.116 %

Finance Agent (v2)Overall. Higher is better. Test setup not documented as identical

Claude Opus 4.853.918 %

GPT-5.6 Luna55.044 %

Key to the marks:better valuebetter, but not conclusivetie

Benchmark scores for Claude Opus 4.8 and GPT-5.6 Luna
Benchmark and task	Claude Opus 4.8	GPT-5.6 Luna	Source
AA-BriefcaseOverall Test details No documented difference in the test setupVersion: v1.1Metric: AA-Briefcase EloScoring: Higher is betterClaude Opus 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	1,321.43 elo	Winner1,342.18 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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	77.7%	Winner83.7%	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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	Winner28.75 points The index can be negative and ranges from -100 to 100 according to Artificial Analysis.	-10.283 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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: Max, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboardClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). 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-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). 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.	Winner48.8% This is the separately published accuracy diagnostic, not the AA-Omniscience Index. AA identifies higher values as better.	42.7% 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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboard GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: Max, Attempts: AA-reported aggregate; diagnostic-specific run count is not stated on the leaderboardClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption. Proportion of incorrect answers among all non-correct responses: incorrect / (incorrect + partial answers + not attempted).	Winner39.3% This is the separately published hallucination-rate diagnostic, not the AA-Omniscience Index. AA identifies lower values as better.	92.6% 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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	Winner41.79 points	37.324 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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	45.6%	Winner50.2%	Artificial Analysis Source details Data as of: 10/19/2026
BrowseCompOverall Test details No documented difference in the test setupVersion: 1,266 tasksMetric: accuracyScoring: Higher is betterCross-vendor comparison table with browsing tools. Exact agent systems differ.	Winner84.3 % Editorial selection from the vendor table, not a complete extract of the comparison cohort.	83.3 % Editorial selection from the vendor table, not a complete extract of the comparison cohort.	OpenAI Source details Data as of: 07/09/2026
Code MigrationOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value47.25 %	44.55 %	Vals AI Source details Data as of: 10/12/2026
CorpFin v2Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value66.706 %	64.219 %	Vals AI Source details Data as of: 08/12/2026
EMBOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value69.373 %	67.116 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	53.918 %	Higher measured value55.044 %	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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	22.8%	Winner24.0%	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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	1,453.47 elo	Winner1,459.2 elo	Artificial Analysis Source details Data as of: 10/19/2026
GPQA DiamondOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value92.424 %	91.666 %	Vals AI Source details Data as of: 09/01/2026
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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value9.583 %	1.25 %	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 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	Winner48.7% Artificial Analysis describes this comparable run as 2,158 text-only questions, not the full 2,500-question dataset.	39.5% 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
Legal Research BenchOverall · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value43.75 %	36.538 %	Vals AI Source details Data as of: 10/12/2026
LegalBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	83.568 %	Higher measured value84.034 %	Vals AI 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 4.8: Snapshot: claude-opus-4-8, Harness: Arena Text, style control GPT-5.6 Luna: Snapshot: gpt-5.6-luna-xhigh, Reasoning: xhigh, Harness: Arena Text, style controlClaude Opus 4.8: Arena rating from 65,172 published pairwise votes.GPT-5.6 Luna: Arena rating from 41,109 published pairwise votes.	Higher measured value1,474.319 95% confidence interval: ±3.846The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,452.566 95% confidence interval: ±4.302The 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: 10/15/2026
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 4.8: Snapshot: claude-opus-4-8, Harness: Arena Vision, style control GPT-5.6 Luna: Snapshot: gpt-5.6-luna-xhigh, Reasoning: xhigh, Harness: Arena Vision, style controlClaude Opus 4.8: Arena rating from 18,852 published pairwise votes.GPT-5.6 Luna: Arena rating from 11,086 published pairwise votes.	Higher measured value1,278.955 95% confidence interval: ±6.534The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,257.662 95% confidence interval: ±7.463The 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 4.8: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 30,000 tokens	Higher measured value53.217 %	42.391 %	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 4.8: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 30,000 tokens	Higher measured value85.755 %	84.391 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value89.585 %	86.036 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value86.59 %	85.029 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value69.912 %	67.29 %	Vals AI Source details Data as of: 09/01/2026
ProgramBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: fully resolved rateScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value1 %	0 %	Vals AI Source details Data as of: 10/12/2026
Public Benefits Bench v1.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value68.133 %	61.164 %	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 4.8: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 30,000 tokens	Higher measured value54.788 %	44.22 %	Vals AI Source details Data as of: 10/08/2026
SkillsBenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Harness: OpenHands, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens	59.227 %	Higher measured value60.447 %	Vals AI Source details Data as of: 09/27/2026
SWE-bench VerifiedOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	88.6 %	Higher measured value93 %	Vals AI Source details Data as of: 09/01/2026
Tax Agent BenchOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value29.999 %	24.815 %	Vals AI Source details Data as of: 10/12/2026
TaxEval v2Overall Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	75.634 %	Higher measured value76.166 %	Vals AI Source details Data as of: 09/01/2026
Terminal-BenchOverall Test details No documented difference in the test setupVersion: 4.0Metric: Task successScoring: Higher is betterClaude Opus 4.8: Snapshot: Claude Opus 4.8 (Max), Reasoning: Max GPT-5.6 Luna: Snapshot: GPT-5.6 Luna (Max), Reasoning: MaxClaude Opus 4.8: Exact AA configuration: Claude Opus 4.8 (Max) (release claude-opus-4-8, row claude-opus-4-8, UUID 992b7b84-5069-4c6a-9295-834252553d50, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.GPT-5.6 Luna: Exact AA configuration: GPT-5.6 Luna (Max) (release gpt-5-6-luna, row gpt-5-6-luna, UUID 426d24c8-49ae-482a-b4a8-20f1c53f21c1, reasoning=true, source reasoning-token configuration budget=0, suffix=Max). This configured value is not measured token consumption.	Winner21.7%	11.6%	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	71.91 %	Higher measured value79.026 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value23.232 %	11.616 %	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 4.8: Compute: max, Harness: Mini-SWE-agent, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Harness: Mini-SWE-agent, Output limit: 128,000 tokens	Higher measured value4.286 %	0 %	Vals AI Source details Data as of: 10/14/2026
Vals IndexOverall Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: weighted index scoreScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value55.1 %	51.687 %	Vals AI Source details Data as of: 10/13/2026
Vals Multimodal IndexOverall Test details Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: weighted index scoreScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value70.889 %	69.058 %	Vals AI Source details Data as of: 08/11/2026
Vibe Code Bench v1.1Overall Test details Test setup not documented as identicalNo direct winnerVersion: 1.1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Harness: OpenHands, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens	Higher measured value82.725 %	77.055 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value40.143 %	36.067 %	Vals AI Source details Data as of: 10/12/2026
CorpFin v2Exact Pages Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value64.336 %	63.52 %	Vals AI Source details Data as of: 08/12/2026
EMBComps Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value72.797 %	54.573 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	48.387 %	Higher measured value52.027 %	Vals AI Source details Data as of: 10/12/2026
GPQA DiamondFew-Shot CoT Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	91.919 %	Higher measured value92.424 %	Vals AI Source details Data as of: 09/01/2026
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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value87.319 %	85.145 %	Vals AI Source details Data as of: 10/12/2026
Legal Research BenchAdministrative / Regulatory · All-pass Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: all-pass rateScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	43.939 %	Higher measured value48.485 %	Vals AI Source details Data as of: 10/12/2026
LegalBenchConclusion Tasks Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value88.937 %	88.864 %	Vals AI 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 4.8: Snapshot: claude-opus-4-8-high, Reasoning: high, Harness: Arena Text, style control GPT-5.6 Luna: Snapshot: gpt-5.6-luna-xhigh, Reasoning: xhigh, Harness: Arena Text, style controlClaude Opus 4.8: Arena rating from 64,100 published pairwise votes.GPT-5.6 Luna: Arena rating from 41,109 published pairwise votes.	Higher measured value1,481.495 95% confidence interval: ±3.868The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,452.566 95% confidence interval: ±4.302The 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: 10/15/2026
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 4.8: Snapshot: claude-opus-4-8-high, Reasoning: high, Harness: Arena Vision, style control GPT-5.6 Luna: Snapshot: gpt-5.6-luna-xhigh, Reasoning: xhigh, Harness: Arena Vision, style controlClaude Opus 4.8: Arena rating from 18,341 published pairwise votes.GPT-5.6 Luna: Arena rating from 11,086 published pairwise votes.	Higher measured value1,285.632 95% confidence interval: ±6.5The Arena score measures human preference rather than a fixed capability test. The confidence interval, Arena category, and model configuration are part of the result.	1,257.662 95% confidence interval: ±7.463The 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
MMLU-ProBiology Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value94.7 %	93.305 %	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 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value79.73 %	75.596 %	Vals AI Source details Data as of: 09/01/2026
ProgramBenchAlmost Resolved Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: almost-resolved rateScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value15.5 %	9 %	Vals AI Source details Data as of: 10/12/2026
SAGEAutomata Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 30,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 30,000 tokens	Higher measured value41.666 %	31.35 %	Vals AI Source details Data as of: 10/08/2026
SkillsBenchNo Skills Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Harness: OpenHands, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Harness: OpenHands, Output limit: 128,000 tokens	43.674 %	Higher measured value45.189 %	Vals AI Source details Data as of: 09/27/2026
SWE-bench Verified<15 min fix Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	92.784 %	Higher measured value96.392 %	Vals AI Source details Data as of: 09/01/2026
Tax Agent BenchControversy & Precedence Analysis Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value13.133 %	10 %	Vals AI Source details Data as of: 10/12/2026
TaxEval v2Correctness Test details Test setup not documented as identicalNo direct winnerVersion: 2Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	64.513 %	Higher measured value66.558 %	Vals AI Source details Data as of: 09/01/2026
Terminal-Bench 2.1Easy Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value100 %	83.333 %	Vals AI Source details Data as of: 09/28/2026
Terminal-Bench 4.0Hardware Test details Test setup not documented as identicalNo direct winnerVersion: 4.0Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value26.667 %	0 %	Vals AI Source details Data as of: 10/13/2026
Terminal-Bench ScienceEarth Sciences Test details Test setup not documented as identicalNo direct winnerVersion: 0.1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Harness: Mini-SWE-agent, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Harness: Mini-SWE-agent, Output limit: 128,000 tokens	No direct winner0 %	No direct winner0 %	Vals AI Source details Data as of: 10/14/2026
Vals IndexCode Migration Test details Test setup not documented as identicalNo direct winnerVersion: 2.1Metric: benchmark scoreScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value46.942 %	44.774 %	Vals AI Source details Data as of: 10/13/2026
Vals Multimodal IndexCorpFin v2 Test details Test setup not documented as identicalNo direct winnerVersion: 1.2Metric: benchmark scoreScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	Higher measured value65.967 %	65.035 %	Vals AI Source details Data as of: 08/11/2026
Code MigrationCOBOL Test details Test setup not documented as identicalNo direct winnerVersion: 1Metric: accuracyScoring: Higher is betterClaude Opus 4.8: Compute: max, Temperature: 1, Output limit: 128,000 tokens GPT-5.6 Luna: Reasoning: max, Output limit: 128,000 tokens	68.571 %	Higher measured value70 %	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: 43b8c5c46830

Claude Opus 4.8

Unavailable in this round

GPT-5.6 Luna

Unavailable in this round

Earlier tests

Pelican on a bicycle

A community classic for free-form SVG drawing.

Claude Opus 4.8

Reasoning (requested): high

Current attempt

Run date: Oct 1, 2026
API provider: openrouter
Requested API model: anthropic/claude-opus-4.8
API model reported in response: anthropic/claude-opus-4.8
Reasoning setting: high
Pinned endpoint: Anthropic | anthropic/claude-4.8-opus-20260528
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: 18 s

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

GPT-5.6 Luna

Reasoning (requested): high

Current attempt

Run date: Sep 7, 2026
API provider: openrouter
Requested API model: openai/gpt-5.6-luna
API model reported in response: openai/gpt-5.6-luna
Reasoning setting: high
Pinned endpoint: OpenAI | openai/gpt-5.6-luna-20260709
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: 49.9 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 4.8 and GPT-5.6 Luna with other leading models using the same data and benchmark logic.

View the complete LLM comparison

Claude Opus 4.8 vs. GPT-5.6 Luna

Change model selection

General specifications

API pricing compared

Benchmark index compared

Shared benchmarks

Capability profile from matched benchmarks

Our tests

SVG short test

Claude Opus 4.8

GPT-5.6 Luna

Pelican on a bicycle

Winning one benchmark is not an overall verdict

Related LLM comparisons