AI benchmark tape
Benchmark Live AI capital allocation same brief, frozen portfolios, real returns AI Positioning Risk-seeking 68.5/100 As of August 30, 2026. Selective risk taking Model Agreement Mixed 5.5 point dispersion Insight Models found the weekly oracle asset The hindsight best asset was Software (IGV) at +5.93%. 4 of 7 models held it, with +17.86% average allocation. The largest allocation came from Claude Fable 5 at +35.00%. Crowded Position S&P 500 24.5% 7 models hold SPY Biggest Shift Real Estate Sector down 20.0 pp vs prior monthly round Open Portfolio Leader GPT-5.5 currently leads at 1.0% interim result, not yet scored Latest Monthly Grok 4.5 led 7.2% vs S&P 3.7% Full Monthly History Grok 4.5 leads score 16.4 across 15/43
Data Prices Aug 28 124/124 audited

One-month rounds

All Available Monthly History

Context across every resolved one-month round. Fair model-vs-model rankings use equal-run comparison sets.

Monthly track

Full-History Monthly Scores

Resolved monthly rounds only.

43 resolved rounds compared3 full-history models ranked6 short-history models separatedNewest included round: CB-2026-07-30-1M
All resolved rounds

Full-History Model Scores

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

Grok 4.3
Gemini 3.1 Pro
GPT-5.5
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 4.3 xAI · 43/43 scored rounds
3.0
Gemini 3.1 Pro Google · 43/43 scored rounds
-2.0
GPT-5.5 OpenAI · 43/43 scored rounds
-8.7
S&P 500 S&P 500 · 43/43 scored rounds
6.6
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
43 resolved rounds compared3 full-history models ranked6 short-history models separatedNewest included round: CB-2026-07-30-1M
Return context

Average Return Details

Average portfolio return across the same finished rounds.

xAI Grok 4.3
0.60%
Google Gemini 3.1 Pro
-0.39%
OpenAI GPT-5.5
-1.74%
S&P S&P 500
1.31%
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.
19.88%
Not ranked yet Short-history models

Shown for transparency; not included in the main ranking until they have all 43 completed rounds.

xAI
Grok 4.5 xAI · short history · 15/43 scored rounds
Score (15/43)
16.4
Avg return
3.57%
OpenAI
GPT-5.6 Sol OpenAI · short history · 13/43 scored rounds
Score (13/43)
12.7
Avg return
2.92%
Anthropic
Claude Opus 5 Anthropic · short history · 5/43 scored rounds
Score (5/43)
12.0
Avg return
3.46%
Anthropic
Claude Fable 5 Anthropic · short history · 21/43 scored rounds
Score (21/43)
9.7
Avg return
2.05%
Anthropic
Claude Opus 4.8 Anthropic · short history · 40/43 scored rounds
Score (40/43)
3.0
Avg return
0.61%
Anthropic
Claude Opus 4.7 Anthropic · short history · 35/43 scored rounds
Score (35/43)
-5.1
Avg return
-0.92%
Short history: models missing resolved rounds are shown but cannot lead the full-history score.
Scope

What This Scorecard Includes

This all-history view can include unequal model histories. Use Benchmark Comparison Sets for the fair headline ranking where every model has the exact same included rounds.

Compare cohorts Comparison sets