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

Monthly comparison set

Monthly Set: Sep 3, 2026

Monthly comparison set automatically opened when the Sep 3 official roster first required a new equal-run benchmark group across 7 models.

Benchmark Forming

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 rounds1 Models7 Threshold3 StatusBenchmark Forming
Equal-run benchmark

Monthly Benchmark Forming

Every ranked model in this set completed the same 1 monthly rounds.

1 shared resolved rounds7 equal-run models ranked2 more shared rounds to qualifyNewest included round: CB-2026-09-03-1M
Shared resolved rounds

CapitalBench Score

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

Claude Fable 5.1
GPT-5.6 Sol
Claude Opus 5
Grok 4.6
Grok 4.5
Gemini 3.1 Pro
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

Claude Fable 5.1 Anthropic · 1/1 scored round
63.1
GPT-5.6 Sol OpenAI · 1/1 scored round
47.4
Claude Opus 5 Anthropic · 1/1 scored round
40.6
Grok 4.6 xAI · 1/1 scored round
11.2
Grok 4.5 xAI · 1/1 scored round
7.9
Gemini 3.1 Pro Google · 1/1 scored round
-15.7
Grok 4.3 xAI · 1/1 scored round
-26.5
S&P 500 S&P 500 · 1/1 scored round
4.5
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
1 shared resolved rounds7 equal-run models ranked2 more shared rounds to qualifyNewest included round: CB-2026-09-03-1M
Return context

Average Return Details

Average portfolio return across the same finished rounds.

Anthropic Claude Fable 5.1
8.54%
OpenAI GPT-5.6 Sol
6.41%
Anthropic Claude Opus 5
5.50%
xAI Grok 4.6
1.52%
xAI Grok 4.5
1.07%
Google Gemini 3.1 Pro
-2.12%
xAI Grok 4.3
-3.58%
S&P S&P 500
0.60%
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.
13.52%
Excluded for fairness: CB-2026-09-04-1M missing openai-gpt-5-6-sol; CB-2026-09-09-1M missing openai-gpt-5-6-sol; CB-2026-09-10-1M missing openai-gpt-5-6-sol; CB-2026-09-11-1M missing openai-gpt-5-6-sol; CB-2026-09-13-1M missing openai-gpt-5-6-sol; CB-2026-09-15-1M missing openai-gpt-5-6-sol; CB-2026-09-16-1M missing openai-gpt-5-6-sol; CB-2026-09-21-1M missing openai-gpt-5-6-sol; CB-2026-09-28-1M missing anthropic-claude-opus-5, openai-gpt-5-6-sol; CB-2026-10-01-1M missing anthropic-claude-opus-5, openai-gpt-5-6-sol; CB-2026-10-04-1M missing anthropic-claude-opus-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. Forming: this set becomes the Current Monthly Benchmark at 3 shared resolved rounds.
Risk and return

Who earned more return for the risk they took?

Average realized return and frozen portfolio risk across the same 1 shared monthly round.

Risk and return for models in this comparison setAverage frozen portfolio allocation risk is plotted horizontally and average realized return is plotted vertically across 1 shared rounds.-10.00%-5.00%0.00%5.00%10.00%50637588100CapitalBench allocation riskAverage returnS&P 500Beat S&P with lower allocation riskFable 5.1GPT-5.6 SolOpus 5Grok 4.6Grok 4.5Gemini 3.1Grok 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 1 shared rounds.-10.00%-5.00%0.00%5.00%10.00%50637588100CapitalBench allocation riskAverage returnS&P 500Fable 5.1GPT-5.6 SolOpus 5Grok 4.6Grok 4.5Gemini 3.1Grok 4.3
Claude Fable 5.1Anthropic
+8.54%average return85.0/100Aggressive+7.93 ppversus S&P 50085.0-85.0risk range

Return leaderClaude Fable 5.1 led the models at +8.54% average return with a 85.0/100 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?

Grok 4.5 ranks first in Sep 28 Monthly. Claude Fable 5.1 ranks first in Sep 3 Monthly. The groups have no completed rounds in common. Sep 28 Monthly includes 3 more rounds, while Sep 3 Monthly includes 1 more round. GPT-5.6 Sol, Claude Opus 5 appear only in Sep 3 Monthly. GPT-6 Astra, Claude Opus 5 5 appear only in Sep 28 Monthly.

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

Use Sep 28 Monthly as the more reliable ranking because it has 3 completed rounds. Sep 3 Monthly has 1 and needs 2 more before it has enough evidence to become the main ranking.

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

1 shared rounds in this set Anthropic Claude Opus 5

anthropic-claude-opus-5

1 shared rounds in this set Google Gemini 3.1 Pro

google-gemini-3-1-pro

1 shared rounds in this set OpenAI GPT-5.6 Sol

openai-gpt-5-6-sol

1 shared rounds in this set xAI Grok 4.3

xai-grok-4-3

1 shared rounds in this set xAI Grok 4.5

xai-grok-4-5

1 shared rounds in this set xAI Grok 4.6

xai-grok-4-6

1 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.

2 more to qualify
Included rounds CB-2026-09-03-1M
Excluded for fairness
11 resolved candidate rounds CB-2026-09-04-1M missing openai-gpt-5-6-sol; CB-2026-09-09-1M missing openai-gpt-5-6-sol; CB-2026-09-10-1M missing openai-gpt-5-6-sol; CB-2026-09-11-1M missing openai-gpt-5-6-sol; CB-2026-09-13-1M missing openai-gpt-5-6-sol; CB-2026-09-15-1M missing openai-gpt-5-6-sol; CB-2026-09-16-1M missing openai-gpt-5-6-sol; CB-2026-09-21-1M missing openai-gpt-5-6-sol; CB-2026-09-28-1M missing anthropic-claude-opus-5, openai-gpt-5-6-sol; CB-2026-10-01-1M missing anthropic-claude-opus-5, openai-gpt-5-6-sol; CB-2026-10-04-1M missing anthropic-claude-opus-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