AI benchmark tape
Benchmark Live AI capital allocation same brief, frozen portfolios, real returns AI Positioning Risk-seeking 68.0/100 As of October 1, 2026. Broad risk seeking Model Agreement Tight 2.5 point dispersion Insight Models found the monthly oracle asset The hindsight best asset was Semiconductors (SMH) at +14.55%. 1 of 7 models held it, with +4.29% average allocation. The largest allocation came from GPT-5.6 Sol at +30.00%. Crowded Position S&P 500 30.4% 6 models hold SPY Biggest Shift S&P 500 up 19.3 pp vs prior weekly round Open Portfolio Leader Claude Fable 5.1 currently leads at 1.9% interim result, not yet scored Latest Weekly Grok 4.5 led -0.8% vs S&P -1.0% Full Weekly History Grok 4.5 leads score 6.0 across 45/72
Data Prices Oct 2 150/150 audited

Weekly comparison set

Weekly Set: Aug 19, 2026

Weekly comparison set automatically opened when the Aug 19 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.

Includes 3 earlier shared rounds completed by all 7 models before this roster began, plus 11 completed since the roster changed.

All comparison sets
Shared rounds14 Models7 Threshold6 StatusQualified Comparison Set
Equal-run benchmark

Weekly Qualified Comparison Set

Every ranked model in this set completed the same 14 weekly rounds.

14 shared resolved rounds7 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-09-02-1W
Shared resolved rounds

CapitalBench Score

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

Grok 4.3
Claude Fable 5
GPT-5.6 Sol
Gemini 3.1 Pro
Grok 4.5
Claude Opus 5
Grok 4.6
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 · 14/14 scored rounds
4.6
Claude Fable 5 Anthropic · 14/14 scored rounds
3.3
GPT-5.6 Sol OpenAI · 14/14 scored rounds
2.4
Gemini 3.1 Pro Google · 14/14 scored rounds
1.6
Grok 4.5 xAI · 14/14 scored rounds
1.4
Claude Opus 5 Anthropic · 14/14 scored rounds
1.0
Grok 4.6 xAI · 14/14 scored rounds
-1.1
S&P 500 S&P 500 · 14/14 scored rounds
-2.2
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
14 shared resolved rounds7 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-09-02-1W
Return context

Average Return Details

Average portfolio return across the same finished rounds.

xAI Grok 4.3
0.63%
Anthropic Claude Fable 5
0.46%
OpenAI GPT-5.6 Sol
0.34%
Google Gemini 3.1 Pro
0.22%
xAI Grok 4.5
0.19%
Anthropic Claude Opus 5
0.14%
xAI Grok 4.6
-0.15%
S&P S&P 500
-0.30%
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.
13.87%
Excluded for fairness: CB-2026-09-03-1W missing anthropic-claude-fable-5; CB-2026-09-04-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-09-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-10-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-11-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-13-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-15-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-16-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-21-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol 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 14 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 14 shared rounds.-1.00%-0.50%0.00%0.50%1.00%5060708090CapitalBench allocation riskAverage returnS&P 500Beat S&P with lower allocation riskGrok 4.3Fable 5GPT-5.6 SolGemini 3.1Grok 4.5Opus 5Grok 4.6
Risk and return for models in this comparison setAverage frozen portfolio allocation risk is plotted horizontally and average realized return is plotted vertically across 14 shared rounds.-1.00%-0.50%0.00%0.50%1.00%5060708090CapitalBench allocation riskAverage returnS&P 500Grok 4.3Fable 5GPT-5.6 SolGemini 3.1Grok 4.5Opus 5Grok 4.6
Grok 4.3xAI
+0.63%average return73.5/100Risk-seeking+0.93 ppversus S&P 50052.9-90.3risk range

Return leaderGrok 4.3 led the models at +0.63% average return with a 73.5/100 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?

Grok 4.5 ranks first in Sep 4 Weekly. Grok 4.3 ranks first in Aug 19 Weekly. The groups have no completed rounds in common. Sep 4 Weekly includes 8 more rounds, while Aug 19 Weekly includes 14 more rounds. Claude Fable 5, GPT-5.6 Sol appear only in Aug 19 Weekly. Claude Fable 5.1, GPT-6 Astra appear only in Sep 4 Weekly.

5models in both 0rounds used by both Changed a lotchange in order Yes top model changed

Sep 4 Weekly is the main published ranking. Aug 19 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

anthropic-claude-fable-5

14 shared rounds in this set Anthropic Claude Opus 5

anthropic-claude-opus-5

14 shared rounds in this set Google Gemini 3.1 Pro

google-gemini-3-1-pro

14 shared rounds in this set OpenAI GPT-5.6 Sol

openai-gpt-5-6-sol

14 shared rounds in this set xAI Grok 4.3

xai-grok-4-3

14 shared rounds in this set xAI Grok 4.5

xai-grok-4-5

14 shared rounds in this set xAI Grok 4.6

xai-grok-4-6

14 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-2026-08-13-1W, CB-2026-08-15-1W, CB-2026-08-18-1W, CB-2026-08-19-1W, CB-2026-08-20-1W, CB-2026-08-21-1W, CB-2026-08-23-1W, CB-2026-08-24-1W, CB-2026-08-25-1W, CB-2026-08-26-1W, CB-2026-08-27-1W, CB-2026-08-30-1W, CB-2026-09-01-1W, CB-2026-09-02-1W
Excluded for fairness
9 resolved candidate rounds CB-2026-09-03-1W missing anthropic-claude-fable-5; CB-2026-09-04-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-09-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-10-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-11-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-13-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-15-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-16-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol; CB-2026-09-21-1W missing anthropic-claude-fable-5, openai-gpt-5-6-sol
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