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
Live dashboard

Live Positions & Returns

Open CapitalBench rounds before final scoring. This page shows frozen model portfolios, current exposure, mark-to-market returns, and the score dates still waiting on final closing prices.

Open live positions See mark-to-market returns Review open tests
Open monthly rounds 19 19 waiting for results
Open weekly rounds 5 5 waiting for results
Frozen live portfolios 174 39 assets currently held
Current risk appetite 75.5/100 As of September 4, 2026 / Risk-seeking
Live tests

Mark-To-Market Portfolio Returns

Open rounds priced to the latest available close. These are interim portfolio returns, not official benchmark scores.

Official results
Live tests

Live Portfolio Returns

Live rounds marked to the latest available close. These are not final scores.

Priced live rounds20 of 24
Latest closeSep 4
Next final scoreSep 8
GPT-5.5OpenAI / 3 open
Portfolio+1.15%S&P 500-0.27%Portfolio Minus S&P 500+1.42%
Claude Fable 5Anthropic / 20 open
Portfolio+0.81%S&P 500+0.16%Portfolio Minus S&P 500+0.64%
Gemini 3.1 ProGoogle / 20 open
Portfolio+0.64%S&P 500+0.16%Portfolio Minus S&P 500+0.47%
Grok 4.3xAI / 20 open
Portfolio+0.58%S&P 500+0.16%Portfolio Minus S&P 500+0.42%
Grok 4.5xAI / 20 open
Portfolio+0.53%S&P 500+0.16%Portfolio Minus S&P 500+0.37%
Claude Opus 4.8Anthropic / 6 open
Portfolio+0.43%S&P 500-0.29%Portfolio Minus S&P 500+0.72%
Claude Opus 5Anthropic / 20 open
Portfolio+0.34%S&P 500+0.16%Portfolio Minus S&P 500+0.17%
GPT-5.6 SolOpenAI / 20 open
Portfolio+0.15%S&P 500+0.16%Portfolio Minus S&P 500-0.02%
Grok 4.6xAI / 17 open
Portfolio+0.05%S&P 500+0.24%Portfolio Minus S&P 500-0.19%
S&P 50020 open tests
S&P 500 return+0.16%CloseSep 4

Interim returns use live rounds only. Completed rounds move to official scored results.

Marked to market from saved entry prices. Official results wait for the scheduled ending close.
Live positions

Current Frozen Model Portfolios

Current asset and model exposure across unresolved monthly and weekly rounds, with each frozen portfolio shown below.

Model profiles
Scope
View
All Open portfolios only
Live AI positioning

S&P 500 (SPY) is the largest live allocation.

21.3% points to S&P 500 (SPY), while US Equity accounts for 65.3% of open portfolios.

Largest assetS&P 500 (SPY)21.3%
Lead categoryUS Equity65.3%
Live rounds24All Open
Portfolios17439 assets held
Category mixClick a category to focus the pick list
Top allocationsAssets with the largest live model allocation
39 assets
29 smaller live allocations30.4%
Monthly model portfolios

CB-2026-09-04-1M

2026-09-08 to 2026-10-08

Waiting for result
Anthropic Claude Fable 5.1
Cybersecurity (CIBR) 35% Regional Banks (KRE) 35% Copper (CPER) 30%
Anthropic Claude Opus 5
Cybersecurity (CIBR) 35% Regional Banks (KRE) 35% S&P 500 (SPY) 30%
Google Gemini 3.1 Pro
Aerospace and Defense (ITA) 35% Industrials Sector (XLI) 35% Consumer Discretionary Sector (XLY) 30%
OpenAI GPT-6 Astra
Copper (CPER) 35% Cybersecurity (CIBR) 35% S&P 500 (SPY) 30%
xAI Grok 4.3
Cybersecurity (CIBR) 35% US Small-Cap Stocks (IWM) 35% Regional Banks (KRE) 30%
xAI Grok 4.5
Cybersecurity (CIBR) 35% Aerospace and Defense (ITA) 35% Industrials Sector (XLI) 30%
xAI Grok 4.6
Cybersecurity (CIBR) 35% Aerospace and Defense (ITA) 35% Industrials Sector (XLI) 30%
Shared top pick Cybersecurity (CIBR) Average across 7 frozen model portfolios.
Top 3 59.3% Spread 5.8 assets
Cybersecurity (CIBR) 30%
Aerospace and Defense (ITA) 15%
Regional Banks (KRE) 14.3%
Industrials Sector (XLI) 13.6%
Monthly Scores Audit packet
Weekly model portfolios

CB-2026-09-04-1W

2026-09-08 to 2026-09-15

Waiting for result
Anthropic Claude Fable 5.1
Software (IGV) 35% Materials Sector (XLB) 35% S&P 500 (SPY) 30%
Anthropic Claude Opus 5
Software (IGV) 35% Materials Sector (XLB) 35% Agriculture Commodities (DBA) 30%
Google Gemini 3.1 Pro
Software (IGV) 35% Agriculture Commodities (DBA) 35% US Dividend Equities (SCHD) 30%
OpenAI GPT-6 Astra
Agriculture Commodities (DBA) 35% S&P 500 (SPY) 65%
xAI Grok 4.3
Agriculture Commodities (DBA) 35% US Dividend Equities (SCHD) 35% Materials Sector (XLB) 30%
xAI Grok 4.5
Software (IGV) 35% Cybersecurity (CIBR) 35% Agriculture Commodities (DBA) 30%
xAI Grok 4.6
Software (IGV) 35% Cybersecurity (CIBR) 35% S&P 500 (SPY) 30%
Shared top pick Software (IGV) Average across 7 frozen model portfolios.
Top 3 66.4% Spread 5.3 assets
Software (IGV) 25%
Agriculture Commodities (DBA) 23.6%
S&P 500 (SPY) 17.9%
Materials Sector (XLB) 14.3%
Weekly Scores Audit packet
Current allocation signal

AI Risk Appetite

Latest monthly and weekly portfolios, equal-weighted by track. This measures current model positioning, not market returns or a trading recommendation.

Historical trend and methodology
Current risk appetite 75.5/100 As of September 4, 2026 / Risk-seeking
As of September 4, 2026 Broad risk seeking CB-2026-09-04-1M and CB-2026-09-04-1W
Monthly strategic 79.8/100 As of September 4, 2026 / Risk-seeking
Weekly tactical 71.1/100 As of September 4, 2026 / Risk-seeking
Change -3.0 Change vs Sep 3 portfolios
Model agreement Tight 4.0 point dispersion
Largest current allocations
Cybersecurity (CIBR) 20.0% S&P 500 (SPY) 13.2% Software (IGV) 12.5% Agriculture Commodities (DBA) 11.8% Aerospace and Defense (ITA) 7.5% Materials Sector (XLB) 7.1%
Regime mix
Broad and cyclical equity 39.3% Growth and technology 32.5% Real assets and inflation 23.6% Defensive equity 4.6%
Model live holdings

Open Portfolios By Model

A compact readout of each model's unresolved CapitalBench exposure. Open the model profile for the full holdings breakdown.

Grok 4.3 xAI Top live holding S&P 500 (SPY) 39.6% across 24 open portfolios Top 3 55.4% Next score Sep 8 Claude Opus 5 Anthropic Top live holding S&P 500 (SPY) 20.4% across 24 open portfolios Top 3 44.8% Next score Sep 8 Grok 4.5 xAI Top live holding S&P 500 (SPY) 11.0% across 24 open portfolios Top 3 29.6% Next score Sep 8 Gemini 3.1 Pro Google Top live holding Utilities Sector (XLU) 11.5% across 24 open portfolios Top 3 27.1% Next score Sep 8 GPT-5.6 Sol OpenAI Top live holding Semiconductors (SMH) 14.1% across 22 open portfolios Top 3 36.8% Next score Sep 8 Grok 4.6 xAI Top live holding S&P 500 (SPY) 53.3% across 21 open portfolios Top 3 67.9% Next score Sep 8 Claude Fable 5 Anthropic Top live holding Semiconductors (SMH) 16.8% across 20 open portfolios Top 3 38.5% Next score Sep 8 Claude Opus 4.8 Anthropic Top live holding S&P 500 (SPY) 62.5% across 6 open portfolios Top 3 80.0% Next score Sep 8 Claude Fable 5.1 Anthropic Top live holding Cybersecurity (CIBR) 17.5% across 4 open portfolios Top 3 48.8% Next score Sep 14 GPT-5.5 OpenAI Top live holding Semiconductors (SMH) 31.7% across 3 open portfolios Top 3 58.3% Next score Sep 8 GPT-6 Astra OpenAI Top live holding S&P 500 (SPY) 47.5% across 2 open portfolios Top 3 82.5% Next score Sep 15
Open tests

Live Monthly And Weekly Tests

These are the open tests you can inspect now. Models already submitted portfolios; official scores wait for final closing prices.

Monthly test

One-month test

Live now

Longer test of AI allocation over one month.

Portfolios locked; scoring pending.

Model portfolios 7 Eligible assets 70 Risk-taking score 79.8/100 Top consensus Cybersecurity (CIBR) 30% average weight
Live positions Asset universe Audit packet
Weekly test

One-week test

Live now

Short-term test of AI positioning over one market week.

Portfolios locked; scoring pending.

Model portfolios 7 Eligible assets 70 Risk-taking score 71.1/100 Top consensus Software (IGV) 25% average weight
Live positions Asset universe Audit packet

Internal IDs and full reproducibility files are inside each audit packet.

Scoring calendar

Current Scoring Calendar

Models have already picked portfolios. Official scores publish only after the market window ends and final closing prices are available.

132 official rounds recorded Internal round and run IDs stay in the public audit trail.
View audit trail
Live data is interim.

Positions are frozen model outputs from open rounds. Returns are marked to the latest available close. Official benchmark scores publish only after the scheduled ending close is collected.