AI benchmark tape
Benchmark Live AI capital allocation same brief, frozen portfolios, real returns AI Positioning Risk-seeking 77.6/100 As of October 28, 2026. Broad risk seeking Model Agreement Mixed 5.7 point dispersion Insight Models missed the monthly oracle asset The hindsight best asset was Ethereum ETF (ETHA) at +152.31%. 0 of 7 models held it, with +0.00% average allocation. Crowded Position S&P 500 36.2% 5 models hold SPY Biggest Shift S&P 500 down 56.7 pp vs prior weekly round Open Portfolio Leader Gemini 3.1 Pro currently leads at 0.7% interim result, not yet scored Latest Weekly Gemini 4 Argon led 0.1% vs S&P -0.6% Full Weekly History Gemini 4 Argon leads score 17.6 across 2/80
Data Prices Nov 3 164/164 audited

Weekly comparison set

Weekly Set: Sep 4, 2026

Weekly comparison set automatically opened when the Sep 4 official roster first required a new equal-run benchmark group across 7 models.

Qualified Comparison Set

Fairness Scope

Every ranked model in this set is scored only on rounds that all 7 listed models completed. If one model misses a resolved round, that round is excluded from this set for everyone.

All comparison sets
Shared rounds8 Models7 Threshold6 StatusQualified Comparison Set
Equal-run benchmark

Weekly Qualified Comparison Set

Every ranked model in this set completed the same 8 weekly rounds.

8 shared resolved rounds7 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-09-21-1W
Shared resolved rounds

CapitalBench Score

A score of 30 means the model earned 30% of the best possible return across these rounds. Calculation

Grok 4.5
Grok 4.6
Gemini 3.1 Pro
Claude Opus 5
Claude Fable 5.1
GPT-6 Astra
Grok 4.3
S&P 500
Max possible What is this? Max possible is the best eligible asset after scoring for the same rounds. It is a hindsight ceiling, not a model portfolio. hindsight best asset

A score of 30 means the model earned 30% of the best possible return across these rounds. Calculation

Grok 4.5 xAI · 8/8 scored rounds
19.0
Grok 4.6 xAI · 8/8 scored rounds
16.1
Gemini 3.1 Pro Google · 8/8 scored rounds
12.1
Claude Opus 5 Anthropic · 8/8 scored rounds
12.0
Claude Fable 5.1 Anthropic · 8/8 scored rounds
5.0
GPT-6 Astra OpenAI · 8/8 scored rounds
1.1
Grok 4.3 xAI · 8/8 scored rounds
-2.9
S&P 500 S&P 500 · 8/8 scored rounds
1.3
Max possible Hindsight ceiling, not a model portfolio
What is this? Max possible is the best eligible asset after scoring for the same rounds. It is a hindsight ceiling, not a model portfolio. 100.0
8 shared resolved rounds7 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-09-21-1W
Return context

Average Return Details

Average portfolio return across the same finished rounds.

xAI Grok 4.5
1.66%
xAI Grok 4.6
1.40%
Google Gemini 3.1 Pro
1.06%
Anthropic Claude Opus 5
1.05%
Anthropic Claude Fable 5.1
0.43%
OpenAI GPT-6 Astra
0.10%
xAI Grok 4.3
-0.25%
S&P S&P 500
0.11%
MAX Max possible What is this? Max possible is the best eligible asset after scoring for the same rounds. It is a hindsight ceiling, not a model portfolio.
8.73%
Excluded for fairness: CB-2026-09-28-1W missing anthropic-claude-opus-5; CB-2026-10-01-1W missing anthropic-claude-opus-5; CB-2026-10-04-1W missing anthropic-claude-opus-5; CB-2026-10-06-1W missing anthropic-claude-opus-5; CB-2026-10-10-1W missing anthropic-claude-opus-5; CB-2026-10-12-1W missing anthropic-claude-opus-5; CB-2026-10-19-1W missing anthropic-claude-opus-5, google-gemini-3-1-pro, xai-grok-4-3; CB-2026-10-22-1W missing anthropic-claude-opus-5, google-gemini-3-1-pro, xai-grok-4-3 Fairness rule: every ranked model completed every included round. A missed round is excluded from this set for everyone.
Risk and return

Who earned more return for the risk they took?

Average realized return and frozen portfolio risk across the same 8 shared weekly rounds.

Risk and return for models in this comparison setAverage frozen portfolio allocation risk is plotted horizontally and average realized return is plotted vertically across 8 shared rounds.-1.00%0.00%1.00%2.00%5060708090CapitalBench allocation riskAverage returnS&P 500Beat S&P with lower allocation riskGrok 4.5Grok 4.6Gemini 3.1Opus 5Fable 5.1GPT-6 AstraGrok 4.3
Risk and return for models in this comparison setAverage frozen portfolio allocation risk is plotted horizontally and average realized return is plotted vertically across 8 shared rounds.-1.00%0.00%1.00%2.00%5060708090CapitalBench allocation riskAverage returnS&P 500Grok 4.5Grok 4.6Gemini 3.1Opus 5Fable 5.1GPT-6 AstraGrok 4.3
Grok 4.5xAI
+1.66%average return70.0/100Risk-seeking+1.55 ppversus S&P 50051.0-81.1risk range

Return leaderGrok 4.5 led the models at +1.66% average return with a 70.0/100 risk score.

Benchmark testNo model beat the S&P 500 while taking the same or less allocation risk in this set.
Compare model groups

How do these results compare?

GPT-6 Astra ranks first in Sep 28 Weekly. Grok 4.5 ranks first in Sep 4 Weekly. The groups have no completed rounds in common. Sep 28 Weekly includes 6 more rounds, while Sep 4 Weekly includes 8 more rounds. Claude Opus 5 appears only in Sep 4 Weekly. Claude Opus 5 5 appears only in Sep 28 Weekly.

6models in both 0rounds used by both Changed a lotchange in order Yes top model changed

Sep 28 Weekly is the main published ranking. Sep 4 Weekly also has enough rounds, so compare them to see whether the results hold across different model groups.

Compare these groups
Roster

Models In This Set

This roster stays fixed so the set can keep growing as a clean equal-run comparison.

Anthropic Claude Fable 5.1

anthropic-claude-fable-5-1

8 shared rounds in this set Anthropic Claude Opus 5

anthropic-claude-opus-5

8 shared rounds in this set Google Gemini 3.1 Pro

google-gemini-3-1-pro

8 shared rounds in this set OpenAI GPT-6 Astra

openai-gpt-6-astra

8 shared rounds in this set xAI Grok 4.3

xai-grok-4-3

8 shared rounds in this set xAI Grok 4.5

xai-grok-4-5

8 shared rounds in this set xAI Grok 4.6

xai-grok-4-6

8 shared rounds in this set
Round audit

Included And Excluded Rounds

Included rounds count toward the score. Excluded rounds are resolved rounds inside this comparison history where at least one set model was missing.

Included rounds CB-2026-09-04-1W, CB-2026-09-09-1W, CB-2026-09-10-1W, CB-2026-09-11-1W, CB-2026-09-13-1W, CB-2026-09-15-1W, CB-2026-09-16-1W, CB-2026-09-21-1W
Excluded for fairness
8 resolved candidate rounds CB-2026-09-28-1W missing anthropic-claude-opus-5; CB-2026-10-01-1W missing anthropic-claude-opus-5; CB-2026-10-04-1W missing anthropic-claude-opus-5; CB-2026-10-06-1W missing anthropic-claude-opus-5; CB-2026-10-10-1W missing anthropic-claude-opus-5; CB-2026-10-12-1W missing anthropic-claude-opus-5; CB-2026-10-19-1W missing anthropic-claude-opus-5, google-gemini-3-1-pro, xai-grok-4-3; CB-2026-10-22-1W missing anthropic-claude-opus-5, google-gemini-3-1-pro, xai-grok-4-3
Calculation

How The Score Is Calculated

CapitalBench Score equals total model return across included shared rounds divided by total max-possible return across those same rounds, multiplied by 100. Max possible is the best eligible asset in each included round in hindsight.

Scoring details