Compare AI Benchmark Sets

Choose two groups

Which results do you want to compare?

A round is one timed market test. A model group uses only rounds completed by every model being ranked together.

First model groupSecond model group

Bottom line

What changed?

Claude Opus 4.8 ranks first in May 28 Weekly. GPT-5.5 ranks first in Jun 9 Weekly. The groups share 5 completed rounds. May 28 Weekly includes 18 more rounds. Claude Fable 5 appears only in Jun 9 Weekly.

Main resultsMay 28 Weekly23 rounds completed by all 5 models Early resultsJun 9 Weekly5 rounds completed by all 6 models

Models in both groups51 only in Jun 9 Weekly

Rounds used by both5May 28 Weekly has 18 more

Did the order change?Changed a lotBased on models in both groups

Did the top model change?Yes1 of 3 stayed in the top three

Model by model

How did each model's result change?

Each group uses all of its completed rounds. Models that appear in only one group are marked clearly.

May 28 Weekly#4

GPT-5.5OpenAIJun 9 Weekly#1Up 3

May 28 WeeklyNot included

Claude Fable 5Only in Jun 9 WeeklyJun 9 Weekly#2Added

May 28 Weekly#2

Grok 4.3xAIJun 9 Weekly#3Down 1

May 28 Weekly#5

Gemini 3.1 ProGoogleJun 9 Weekly#4Up 1

May 28 Weekly#1

Claude Opus 4.8AnthropicJun 9 Weekly#5Down 4

May 28 Weekly#3

Claude Opus 4.7AnthropicJun 9 Weekly#6Down 3

Fair comparison

Did both groups use the same rounds?

5 completed rounds are used by both groups. May 28 Weekly also uses 18 other rounds. Different rounds can change scores and ranks.

5 used by both18 only in May 28 Weekly0 only in Jun 9 Weekly

GPT-5.5Overall score on 5 shared rounds17.8Overall score on 18 other May 28 Weekly rounds-28.6Overall score on all 23 May 28 Weekly rounds-17.3

Grok 4.3Overall score on 5 shared rounds8.5Overall score on 18 other May 28 Weekly rounds-12.0Overall score on all 23 May 28 Weekly rounds-7.0

Gemini 3.1 ProOverall score on 5 shared rounds7.9Overall score on 18 other May 28 Weekly rounds-28.5Overall score on all 23 May 28 Weekly rounds-19.7

Claude Opus 4.8Overall score on 5 shared rounds7.2Overall score on 18 other May 28 Weekly rounds-8.3Overall score on all 23 May 28 Weekly rounds-4.5

Claude Opus 4.7Overall score on 5 shared rounds5.7Overall score on 18 other May 28 Weekly rounds-12.9Overall score on all 23 May 28 Weekly rounds-8.4

See exactly which rounds were used

Used by both: CB-2026-06-09-1W, CB-2026-06-12-1W, CB-2026-06-13-1W, CB-2026-07-01-1W, CB-2026-07-02-1W

Only in May 28 Weekly: CB-2026-05-28-1W, CB-2026-05-29-1W, CB-2026-06-01-1W, CB-2026-06-02-1W, CB-2026-06-03-1W, CB-2026-06-05-1W, CB-2026-06-08-1W, CB-2026-06-15-1W, CB-2026-06-16-1W, CB-2026-06-17-1W, CB-2026-06-18-1W, CB-2026-06-22-1W, CB-2026-06-23-1W, CB-2026-06-24-1W, CB-2026-06-25-1W, CB-2026-06-26-1W, CB-2026-06-29-1W, CB-2026-06-30-1W

Only in Jun 9 Weekly: None

Why were 11 rounds left out of Jun 9 Weekly?

All 11 rounds were left out because Claude Fable 5 had no recorded result. To keep the ranking fair, those rounds were left out for every model in Jun 9 Weekly.

Claude Fable 5 had no result: Jun 15, 2026, Jun 16, 2026, Jun 17, 2026, Jun 18, 2026, Jun 22, 2026, Jun 23, 2026, Jun 24, 2026, Jun 25, 2026, Jun 26, 2026, Jun 29, 2026, Jun 30, 2026

What this means

Which results should you rely on?

Use May 28 Weekly as the more reliable ranking because it has 23 completed rounds. Jun 9 Weekly has 5 and needs 1 more before it has enough evidence to become the main ranking.

Open May 28 Weekly Open Jun 9 Weekly

All results

What are all the numbers?

Model	Included in	May 28 Weekly rank	Jun 9 Weekly rank	Change	May 28 Weekly overall score	Jun 9 Weekly overall score	Overall score change
GPT-5.5OpenAI	Both groups	#4	#1	Up 3	-17.3	17.8	+35.1
Claude Fable 5Anthropic	Only Jun 9 Weekly	—	#2	Added	n/a	16.3	n/a
Grok 4.3xAI	Both groups	#2	#3	Down 1	-7.0	8.5	+15.5
Gemini 3.1 ProGoogle	Both groups	#5	#4	Up 1	-19.7	7.9	+27.6
Claude Opus 4.8Anthropic	Both groups	#1	#5	Down 4	-4.5	7.2	+11.8
Claude Opus 4.7Anthropic	Both groups	#3	#6	Down 3	-8.4	5.7	+14.0