ARC-AGI-2

Accessible data for ARC-AGI-2 leaderboard
Model	Rank	Model ID	Provider	Score	Unit
GPT-5.6 Sol	#1	gpt-5-6-sol	OpenAI	92.5	percent
Claude Opus 5	#2	claude-opus-5	Anthropic	90.4	percent
GPT-5.5	#3	gpt-5-5	OpenAI	85	percent
GPT-5.6 Terra	#4	gpt-5-6-terra	OpenAI	83.9	percent
GPT-5.4	#5	gpt-5-4	OpenAI	83.3	percent
Gemini 3.1 Pro Preview	#6	gemini-3-1-pro-preview	Google	77.1	percent
Claude Opus 4.7	#7	claude-opus-4-7	Anthropic	75.8	percent
Gemini 3.5 Flash	#8	gemini-3-5-flash	Google	72.1	percent
Claude Opus 4.8	#9	claude-opus-4-8	Anthropic	72.08	percent
GPT-5.6 Luna	#10	gpt-5-6-luna	OpenAI	59.54	percent
GPT-5.2	#11	gpt-5-2	OpenAI	54.2	percent
Grok 4.20	#12	grok-4-20	xAI	53.3	percent
Grok 4.5	#13	grok-4-5	xAI	52.64	percent
Muse Spark	#14	muse-spark	Meta	42.5	percent
Inkling-Small	#15	inkling-small	Thinking Machines Lab	40.1	percent
Claude Sonnet 4.5	#16	claude-sonnet-4-5	Anthropic	13.6	percent

ARC-AGI-2 leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	Source carrier	92.5%
#2	Claude Opus 5 claude-opus-5	Anthropic	closed	Source carrier	90.4%
#3	GPT-5.5 gpt-5.5	OpenAI	closed	Source carrier	85%
#4	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	Source carrier	83.9%
#5	GPT-5.4 gpt-5.4	OpenAI	closed	Source carrier	83.3%
#6	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	Public reference	77.1%
#7	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	Source carrier	75.8%
#8	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	Score input	72.1%
#9	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	Source carrier	72.1%
#10	GPT-5.6 Luna gpt-5.6-luna	OpenAI	closed	Source carrier	59.5%
#11	GPT-5.2 gpt-5.2	OpenAI	closed	Source carrier	54.2%
#12	Grok 4.20 grok-4.20	xAI	closed	Source carrier	53.3%
#13	Grok 4.5 grok-4.5	xAI	closed	Source carrier	52.6%
#14	Muse Spark muse-spark	Meta	closed	Source carrier	42.5%
#15	Inkling-Small thinkingmachines/inkling-small	Thinking Machines Lab	open	Source carrier	40.1%
#16	Claude Sonnet 4.5 claude-sonnet-4-5	Anthropic	closed	Source carrier	13.6%

Rank

Model

Provider

License

Evidence use

Score

GPT-5.6 Sol

gpt-5.6-sol

OpenAI

closed

Source carrier

92.5%

Claude Opus 5

claude-opus-5

Anthropic

closed

Source carrier

90.4%

GPT-5.5

gpt-5.5

OpenAI

closed

Source carrier

85%

GPT-5.6 Terra

gpt-5.6-terra

OpenAI

closed

Source carrier

83.9%

GPT-5.4

gpt-5.4

OpenAI

closed

Source carrier

83.3%

Gemini 3.1 Pro Preview

gemini-3.1-pro-preview

Google

closed

Public reference

77.1%

Claude Opus 4.7

claude-opus-4-7

Anthropic

closed

Source carrier

75.8%

Gemini 3.5 Flash

gemini-3.5-flash

Google

closed

Score input

72.1%

Claude Opus 4.8

claude-opus-4-8

Anthropic

closed

Source carrier

72.1%

#10

GPT-5.6 Luna

gpt-5.6-luna

OpenAI

closed

Source carrier

59.5%

#11

GPT-5.2

gpt-5.2

OpenAI

closed

Source carrier

54.2%

#12

Grok 4.20

grok-4.20

xAI

closed

Source carrier

53.3%

#13

Grok 4.5

grok-4.5

xAI

closed

Source carrier

52.6%

#14

Muse Spark

muse-spark

Every distinct dated source-published row for ARC-AGI-2, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	92.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	90.4%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	85%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	83.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4 Pro; bulk export does not retain a complete upstream harness configuration. gpt-5-4-pro-default-medium	GPT-5.4	83.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	75.8%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	74.0%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.8	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	59.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	53.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	52.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	52.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	42.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Inkling-Small; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling-Small	40.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Sonnet 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.5	13.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	92.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	90.4%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	85%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	83.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4 Pro; bulk export does not retain a complete upstream harness configuration. gpt-5-4-pro-default-medium	GPT-5.4	83.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	75.8%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	74.0%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.8	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	59.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	53.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	52.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	52.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	42.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Sonnet 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.5	13.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	85%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.4 Pro; bulk export does not retain a complete upstream harness configuration. gpt-5-4-pro-default-medium	GPT-5.4	83.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	75.8%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	53.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	52.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	42.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Sonnet 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.5	13.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	GPT-5.5	85%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Gemini 3.1 Pro Preview	77.1%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. gpt-5-2-pro-unspecified	GPT-5.2	54.2%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Muse Spark	42.5%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	GPT-5.5	84.6%	2Provider evaluation harness	Public reference · not scored provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	Gemini 3.1 Pro Preview	77.1%	2Provider evaluation harness	Public reference · not scored provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2Provider evaluation harness	Score input provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15

What does ARC-AGI-2 measure?

A second-generation abstract reasoning benchmark based on novel grid transformation tasks.

Which model scores highest on ARC-AGI-2?

GPT-5.6 Sol by OpenAI currently leads with 92.5%.

How many models are evaluated on ARC-AGI-2?

16 canonical model products have a qualifying score, backed by 47 distinct dated source rows.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 2.5% of its capability category. GPT-5.6 Sol is overall #4.

Accessible data for ARC-AGI-2 leaderboard
Model	Rank	Model ID	Provider	Score	Unit
GPT-5.6 Sol	#1	gpt-5-6-sol	OpenAI	92.5	percent
Claude Opus 5	#2	claude-opus-5	Anthropic	90.4	percent
GPT-5.5	#3	gpt-5-5	OpenAI	85	percent
GPT-5.6 Terra	#4	gpt-5-6-terra	OpenAI	83.9	percent
GPT-5.4	#5	gpt-5-4	OpenAI	83.3	percent
Gemini 3.1 Pro Preview	#6	gemini-3-1-pro-preview	Google	77.1	percent
Claude Opus 4.7	#7	claude-opus-4-7	Anthropic	75.8	percent
Gemini 3.5 Flash	#8	gemini-3-5-flash	Google	72.1	percent
Claude Opus 4.8	#9	claude-opus-4-8	Anthropic	72.08	percent
GPT-5.6 Luna	#10	gpt-5-6-luna	OpenAI	59.54	percent
GPT-5.2	#11	gpt-5-2	OpenAI	54.2	percent
Grok 4.20	#12	grok-4-20	xAI	53.3	percent
Grok 4.5	#13	grok-4-5	xAI	52.64	percent
Muse Spark	#14	muse-spark	Meta	42.5	percent
Inkling-Small	#15	inkling-small	Thinking Machines Lab	40.1	percent
Claude Sonnet 4.5	#16	claude-sonnet-4-5	Anthropic	13.6	percent

ARC-AGI-2 leaderboard ranked by score
Rank	Model	Provider	License	Evidence use	Score
#1	GPT-5.6 Sol gpt-5.6-sol	OpenAI	closed	Source carrier	92.5%
#2	Claude Opus 5 claude-opus-5	Anthropic	closed	Source carrier	90.4%
#3	GPT-5.5 gpt-5.5	OpenAI	closed	Source carrier	85%
#4	GPT-5.6 Terra gpt-5.6-terra	OpenAI	closed	Source carrier	83.9%
#5	GPT-5.4 gpt-5.4	OpenAI	closed	Source carrier	83.3%
#6	Gemini 3.1 Pro Preview gemini-3.1-pro-preview	Google	closed	Public reference	77.1%
#7	Claude Opus 4.7 claude-opus-4-7	Anthropic	closed	Source carrier	75.8%
#8	Gemini 3.5 Flash gemini-3.5-flash	Google	closed	Score input	72.1%
#9	Claude Opus 4.8 claude-opus-4-8	Anthropic	closed	Source carrier	72.1%
#10	GPT-5.6 Luna gpt-5.6-luna	OpenAI	closed	Source carrier	59.5%
#11	GPT-5.2 gpt-5.2	OpenAI	closed	Source carrier	54.2%
#12	Grok 4.20 grok-4.20	xAI	closed	Source carrier	53.3%
#13	Grok 4.5 grok-4.5	xAI	closed	Source carrier	52.6%
#14	Muse Spark muse-spark	Meta	closed	Source carrier	42.5%
#15	Inkling-Small thinkingmachines/inkling-small	Thinking Machines Lab	open	Source carrier	40.1%
#16	Claude Sonnet 4.5 claude-sonnet-4-5	Anthropic	closed	Source carrier	13.6%

Rank

Model

Provider

License

Evidence use

Score

GPT-5.6 Sol

gpt-5.6-sol

OpenAI

closed

Source carrier

92.5%

Claude Opus 5

claude-opus-5

Anthropic

closed

Source carrier

90.4%

GPT-5.5

gpt-5.5

OpenAI

closed

Source carrier

85%

GPT-5.6 Terra

gpt-5.6-terra

OpenAI

closed

Source carrier

83.9%

GPT-5.4

gpt-5.4

OpenAI

closed

Source carrier

83.3%

Gemini 3.1 Pro Preview

gemini-3.1-pro-preview

Google

closed

Public reference

77.1%

Claude Opus 4.7

claude-opus-4-7

Anthropic

closed

Source carrier

75.8%

Gemini 3.5 Flash

gemini-3.5-flash

Google

closed

Score input

72.1%

Claude Opus 4.8

claude-opus-4-8

Anthropic

closed

Source carrier

72.1%

#10

GPT-5.6 Luna

gpt-5.6-luna

OpenAI

closed

Source carrier

59.5%

#11

GPT-5.2

gpt-5.2

OpenAI

closed

Source carrier

54.2%

#12

Grok 4.20

grok-4.20

xAI

closed

Source carrier

53.3%

#13

Grok 4.5

grok-4.5

xAI

closed

Source carrier

52.6%

#14

Muse Spark

muse-spark

Every distinct dated source-published row for ARC-AGI-2, including exact model configuration labels
Exact published model / configuration	Canonical product	Score	Version / system	Evidence	Source / observed
Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	92.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	90.4%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	85%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	83.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4 Pro; bulk export does not retain a complete upstream harness configuration. gpt-5-4-pro-default-medium	GPT-5.4	83.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	75.8%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	74.0%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.8	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	59.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	53.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	52.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	52.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	42.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Inkling-Small; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Inkling-Small	40.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant Claude Sonnet 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.5	13.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-08-01 ↗Observed 2026-08-01 · checked 2026-08-01
Exact BenchLM registry variant GPT-5.6 Sol; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Sol	92.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 5	90.4%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	85%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.6 Terra; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Terra	83.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4 Pro; bulk export does not retain a complete upstream harness configuration. gpt-5-4-pro-default-medium	GPT-5.4	83.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	75.8%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.4; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.4	74.0%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Opus 4.8; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Opus 4.8	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.6 Luna; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.6 Luna	59.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	53.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	52.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Grok 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.5	52.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	42.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant Claude Sonnet 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.5	13.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 2026-07-27 ↗Observed 2026-07-27 · checked 2026-07-27
Exact BenchLM registry variant GPT-5.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.5	85%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.4 Pro; bulk export does not retain a complete upstream harness configuration. gpt-5-4-pro-default-medium	GPT-5.4	83.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Opus 4.7 (Adaptive); bulk export does not retain a complete upstream harness configuration. claude-opus-4-7-max	Claude Opus 4.7	75.8%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Gemini 3.5 Flash; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Grok 4.20; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Grok 4.20	53.3%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant GPT-5.2; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	GPT-5.2	52.9%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Muse Spark; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Muse Spark	42.5%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact BenchLM registry variant Claude Sonnet 4.5; bulk export does not retain a complete upstream harness configuration. source label without registered configuration ID	Claude Sonnet 4.5	13.6%	2025Source-native system	Source carrier · no additional scoring weight source-checked	BenchLM public datasets — 21 July 2026 ↗Observed 2026-07-21 · checked 2026-07-21
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	GPT-5.5	85%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Gemini 3.1 Pro Preview	77.1%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. gpt-5-2-pro-unspecified	GPT-5.2	54.2%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant as listed on the BenchLM public benchmark leaderboard page. source label without registered configuration ID	Muse Spark	42.5%	2BenchLM aggregated public evaluation	Source carrier · no additional scoring weight source-checked	BenchLM public benchmark leaderboards ↗Observed 2026-07-15 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	GPT-5.5	84.6%	2Provider evaluation harness	Public reference · not scored provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	Gemini 3.1 Pro Preview	77.1%	2Provider evaluation harness	Public reference · not scored provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15
Exact model variant and provider evaluation configuration stated in the Gemini 3.5 Flash model card; single-attempt where specified. source label without registered configuration ID	Gemini 3.5 Flash	72.1%	2Provider evaluation harness	Score input provider-reported	Gemini 3.5 Flash model card ↗Observed 2026-05-19 · checked 2026-07-15

What does ARC-AGI-2 measure?

A second-generation abstract reasoning benchmark based on novel grid transformation tasks.

Which model scores highest on ARC-AGI-2?

GPT-5.6 Sol by OpenAI currently leads with 92.5%.

How many models are evaluated on ARC-AGI-2?

16 canonical model products have a qualifying score, backed by 47 distinct dated source rows.

Does this affect overall Lumina rank?

Yes. This family is ranking-weighted at 2.5% of its capability category. GPT-5.6 Sol is overall #4.

Navigation

ARC-AGI-2

Leaderboard · 16 models

GPT-5.6 Sol

Claude Opus 5

GPT-5.5

ARC-AGI-2 leaderboard

Crop ARC-AGI-2 leaderboard

Benchmark score table (16 models)

Crop ARC-AGI-2 benchmark table

Crop ARC-AGI-2 published source rows

Compare top models

Definition and scoring

Quick answers

What does ARC-AGI-2 measure?

Which model scores highest on ARC-AGI-2?

How many models are evaluated on ARC-AGI-2?

Does this affect overall Lumina rank?

ARC-AGI-2

Leaderboard · 16 models

GPT-5.6 Sol

Claude Opus 5

GPT-5.5

ARC-AGI-2 leaderboard

Crop ARC-AGI-2 leaderboard

Benchmark score table (16 models)

Crop ARC-AGI-2 benchmark table

Crop ARC-AGI-2 published source rows

Compare top models

Definition and scoring

Quick answers

What does ARC-AGI-2 measure?

Which model scores highest on ARC-AGI-2?

How many models are evaluated on ARC-AGI-2?

Does this affect overall Lumina rank?

Leaderboard · 16 models

GPT-5.6 Sol

Claude Opus 5

GPT-5.5

Benchmark score table (16 models)

Compare top models

Definition and scoring

Quick answers

What does ARC-AGI-2 measure?

Which model scores highest on ARC-AGI-2?

How many models are evaluated on ARC-AGI-2?

Does this affect overall Lumina rank?

More Reasoning benchmarks

Leaderboard · 16 models

GPT-5.6 Sol

Claude Opus 5

GPT-5.5

Benchmark score table (16 models)

Compare top models

Definition and scoring

Quick answers

What does ARC-AGI-2 measure?

Which model scores highest on ARC-AGI-2?

How many models are evaluated on ARC-AGI-2?

Does this affect overall Lumina rank?

More Reasoning benchmarks