AI benchmark tape
Benchmark Live AI capital allocation same brief, frozen portfolios, real returns AI Positioning Risk-seeking 75.5/100 As of September 4, 2026. Broad risk seeking Model Agreement Tight 4.0 point dispersion Insight Models found the weekly oracle asset The hindsight best asset was Crude Oil (USO) at +9.45%. 1 of 7 models held it, with +5.00% average allocation. The largest allocation came from Gemini 3.1 Pro at +35.00%. Crowded Position S&P 500 21.3% 9 models hold SPY Biggest Shift Agriculture Commodities up 23.6 pp vs prior weekly round Open Portfolio Leader GPT-5.5 currently leads at 1.2% interim result, not yet scored Latest Weekly Gemini 3.1 Pro led 3.5% vs S&P 0.1% Full Weekly History GPT-5.6 Sol leads score 8.5 across 31/60
Data Prices Sep 4 132/132 audited

Monthly comparison set

Monthly Set: Jul 24, 2026

Monthly comparison set automatically opened when the Jul 24 official roster first required a new equal-run benchmark group across 8 models.

Current Benchmark

Fairness Scope

Every ranked model in this set is scored only on rounds that all 8 listed models completed. If one model misses a resolved round, that round is excluded from this set for everyone.

All comparison sets
Shared rounds8 Models8 Threshold3 StatusCurrent Benchmark
Equal-run benchmark

Current Monthly Benchmark

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

8 shared resolved rounds8 equal-run models rankedQualified at 3+ shared roundsNewest included round: CB-2026-08-05-1M
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
Claude Fable 5
Gemini 3.1 Pro
GPT-5.5
GPT-5.6 Sol
Claude Opus 5
Claude Opus 4.8
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
13.0
Claude Fable 5 Anthropic · 8/8 scored rounds
11.9
Gemini 3.1 Pro Google · 8/8 scored rounds
10.7
GPT-5.5 OpenAI · 8/8 scored rounds
10.5
GPT-5.6 Sol OpenAI · 8/8 scored rounds
10.0
Claude Opus 5 Anthropic · 8/8 scored rounds
9.2
Claude Opus 4.8 Anthropic · 8/8 scored rounds
6.8
Grok 4.3 xAI · 8/8 scored rounds
5.6
S&P 500 S&P 500 · 8/8 scored rounds
9.9
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 rounds8 equal-run models rankedQualified at 3+ shared roundsNewest included round: CB-2026-08-05-1M
Return context

Average Return Details

Average portfolio return across the same finished rounds.

xAI Grok 4.5
3.82%
Anthropic Claude Fable 5
3.49%
Google Gemini 3.1 Pro
3.16%
OpenAI GPT-5.5
3.08%
OpenAI GPT-5.6 Sol
2.95%
Anthropic Claude Opus 5
2.72%
Anthropic Claude Opus 4.8
2.02%
xAI Grok 4.3
1.65%
S&P S&P 500
2.91%
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.
29.43%
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 monthly 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.-2.00%0.00%2.00%4.00%6.00%5060708090CapitalBench allocation riskAverage returnS&P 500Beat S&P with lower allocation riskGrok 4.5Fable 5Gemini 3.1GPT-5.5GPT-5.6 SolOpus 5Opus 4.8Grok 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.-2.00%0.00%2.00%4.00%6.00%5060708090CapitalBench allocation riskAverage returnS&P 500Grok 4.5Fable 5Gemini 3.1GPT-5.5GPT-5.6 SolOpus 5Opus 4.8Grok 4.3
Grok 4.5xAI
+3.82%average return71.1/100Risk-seeking+0.91 ppversus S&P 50062.3-87.8risk range

Return leaderGrok 4.5 led the models at +3.82% average return with a 71.1/100 risk score.

Benchmark testGemini 3.1 Pro beat the S&P 500 while taking no more allocation risk.
Compare model groups

How do these results compare?

Grok 4.5 ranks first in both groups. The groups share 8 completed rounds. Jul 21 Monthly includes 8 more rounds. Claude Opus 5 appears only in Jul 24 Monthly.

7models in both 8rounds used by both Changed a littlechange in order No top model changed

Jul 24 Monthly is the main published ranking. Jul 21 Monthly 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

anthropic-claude-fable-5

8 shared rounds in this set Anthropic Claude Opus 4.8

anthropic-claude-opus-4-8

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-5.5

openai-gpt-5-5

8 shared rounds in this set OpenAI GPT-5.6 Sol

openai-gpt-5-6-sol

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
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-2026-07-24-1M, CB-2026-07-27-1M, CB-2026-07-28-1M, CB-2026-07-29-1M, CB-2026-07-30-1M, CB-2026-07-31-1M, CB-2026-08-04-1M, CB-2026-08-05-1M
Excluded for fairness None
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