Results

Benchmark Results

AI model portfolios are scored in separate weekly and monthly tracks using the same rules, frozen portfolios, and real public-market prices.

Latest weekly Comparison sets Market Regime Benchmark All weekly history Combined results Audit and data

Benchmark status

129 completed 15 live

Weekly and monthly tracks are scored separately.

Current weekly benchmark leader

Grok 4.5 20.6 CapitalBench Score

Latest scored CB-2026-08-25-1M Latest live CB-2026-09-21-1W / CB-2026-09-21-1M Models 12 Universe 70 options

Completed
Live
Scored

Latest weekly result Comparison sets All rounds

Results insights

What Resolved Results Reveal

Signals generated from scored rounds, market environments, oracle comparisons, benchmark difficulty, and model confidence behavior.

Market EnvironmentAs of Sep 25

Monthly market environments12 resolved rounds2 models

Monthly model leadership changes with the S&P 500 environment

Grok 4.3 leads down environments at -1.22% across 6 tests; Grok 4.5 leads up environments at +4.11% across 6 tests.

Market environments group resolved rounds by the S&P 500 return over the same weekly or monthly window. Models are compared only on shared rounds; high confidence requires at least six observations and stable leadership.

Medium confidenceMath: deterministicData through Sep 25, 2026

Down Leader Average Return: -1.22%
Down Shared Rounds: 6
Down Leader Stability: 1.00

Consensus PerformanceAug 25-Sep 25

Monthly resultCB-2026-08-25-1M7 models

AI consensus portfolio scored -23.4 versus the oracle

If the monthly model allocations were averaged into one consensus portfolio, it returned -3.86% versus +0.69% for the S&P 500 and +16.47% for the hindsight best asset.

Consensus means the average of model allocations in the same round. CapitalBench Score compares that return with the hindsight-best eligible asset for that exact scoring window.

High confidenceMath: deterministicData through Sep 25, 2026

Consensus Portfolio Return: -3.86%
Average Model Return: -3.86%
Consensus Capitalbench Score: -23.4

Benchmark DifficultyAug 25-Sep 25

Monthly resultCB-2026-08-25-1M7 models

Monthly round had +26.43% asset dispersion

The best scored asset returned +16.47%, the worst returned -9.96%, and +32.86% of the universe was positive. The S&P 500 ranked 21 out of 70 options.

Asset dispersion is the gap between the best and worst eligible assets in the same round. Wider dispersion makes missed allocation choices more costly.

High confidenceMath: deterministicData through Sep 25, 2026

Oracle Return: +16.5%
Worst Asset Return: -9.96%
Positive Universe Share: +32.9%

All insights

Weekly track

Grok 4.5 Leads

CapitalBench Score leader inside the featured equal-run comparison set.

7 scored

Grok 4.5 xAI

CapitalBench Score 20.6 Avg return leader 2.02% Shared rounds 7 Timeline One market week

Latest result Comparison sets All history Audit packet

Monthly track

Claude Fable 5 Leads

CapitalBench Score leader inside the featured equal-run comparison set.

9 scored

Claude Fable 5 Anthropic

CapitalBench Score 8.4 Avg return leader 1.94% Shared rounds 9 Timeline One market month

Latest result Comparison sets All history Audit packet

Completed results

Current Benchmark Scores

These are equal-run comparison sets. Every ranked model completed every included round, and missed rounds are excluded from the set for everyone.

Open comparison sets

Equal-run benchmark

Current Weekly Benchmark

Every ranked model in this set completed the same 7 weekly rounds.

7 shared resolved rounds7 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-09-16-1W

Shared resolved rounds

CapitalBench Score

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

Grok 4.5

Grok 4.6

Gemini 3.1 Pro

Claude Opus 5

Claude Fable 5.1

GPT-6 Astra

Grok 4.3

S&P 500

Max possible What is this? 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 · 7/7 scored rounds

20.6

Grok 4.6 xAI · 7/7 scored rounds

17.9

Gemini 3.1 Pro Google · 7/7 scored rounds

16.6

Claude Opus 5 Anthropic · 7/7 scored rounds

14.2

Claude Fable 5.1 Anthropic · 7/7 scored rounds

7.6

GPT-6 Astra OpenAI · 7/7 scored rounds

3.9

Grok 4.3 xAI · 7/7 scored rounds

-1.4

S&P 500 S&P 500 · 7/7 scored rounds

2.8

Max possible Hindsight ceiling, not a model portfolio

What is this? 100.0

7 shared resolved rounds7 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-09-16-1W

Return context

Average Return Details

Average portfolio return across the same finished rounds.

Grok 4.5

2.02%

Grok 4.6

1.75%

Gemini 3.1 Pro

1.62%

Claude Opus 5

1.39%

Claude Fable 5.1

0.75%

GPT-6 Astra

0.38%

Grok 4.3

-0.14%

S&P S&P 500

0.27%

MAX Max possible What is this?

9.78%

Fairness rule: every ranked model completed every included round. A missed round is excluded from this set for everyone.

Equal-run benchmark

Current Monthly Benchmark

Every ranked model in this set completed the same 9 monthly rounds.

9 shared resolved rounds7 equal-run models rankedQualified at 3+ shared roundsNewest included round: CB-2026-08-25-1M

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

GPT-5.6 Sol

Claude Opus 5

Grok 4.3

Grok 4.5

Grok 4.6

Gemini 3.1 Pro

S&P 500

Max possible What is this? hindsight best asset

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

Claude Fable 5 Anthropic · 9/9 scored rounds

8.4

GPT-5.6 Sol OpenAI · 9/9 scored rounds

2.8

Claude Opus 5 Anthropic · 9/9 scored rounds

0.3

Grok 4.3 xAI · 9/9 scored rounds

-0.2

Grok 4.5 xAI · 9/9 scored rounds

-2.1

Grok 4.6 xAI · 9/9 scored rounds

-4.3

Gemini 3.1 Pro Google · 9/9 scored rounds

-6.8

S&P 500 S&P 500 · 9/9 scored rounds

-0.4

Max possible Hindsight ceiling, not a model portfolio

What is this? 100.0

9 shared resolved rounds7 equal-run models rankedQualified at 3+ shared roundsNewest included round: CB-2026-08-25-1M

Return context

Average Return Details

Average portfolio return across the same finished rounds.

Claude Fable 5

1.94%

GPT-5.6 Sol

0.65%

Claude Opus 5

0.07%

Grok 4.3

-0.05%

Grok 4.5

-0.49%

Grok 4.6

-0.99%

Gemini 3.1 Pro

-1.59%

S&P S&P 500

-0.10%

MAX Max possible What is this?

23.28%

Fairness rule: every ranked model completed every included round. A missed round is excluded from this set for everyone.

Evidence context

Evidence Behind The Current Benchmarks

Generated from completed rounds, benchmark-set rules, protocols, and available baselines so every score carries its own context.

Score methodology

Weekly benchmarkMore established

Evidence levelMore establishedWeekly evidence has enough completed rounds for stronger pattern reads, while still needing ongoing live validation.

Weekly evidence71 resolved rounds / 468 model resultsCurrent threshold met at 6+ rounds

Equal-run comparison7 models on the same 7 roundsRanked models are compared only on rounds every model in the roster completed.

ProtocolPortfolio-onlyCompleted rounds use constrained multi-asset portfolios.

Score scaleOracle-relative100 means matching the hindsight best asset in the same scored window.

Baselines shownS&P 500, Cash, Oracle, AI consensus portfolioPractical references are shown beside the impossible hindsight ceiling when available.

Use this as benchmark evidence, not an investable strategy result. More resolved rounds are needed before making strong performance claims.

Monthly benchmarkMore established

Evidence levelMore establishedMonthly evidence has enough completed rounds for stronger pattern reads, while still needing ongoing live validation.

Monthly evidence58 resolved rounds / 372 model resultsCurrent threshold met at 3+ rounds

Equal-run comparison7 models on the same 9 roundsRanked models are compared only on rounds every model in the roster completed.

ProtocolMixed protocolCompleted history includes 57 portfolio, 1 single-pick, and 0 unlabelled rounds.

Score scaleOracle-relative100 means matching the hindsight best asset in the same scored window.

Baselines shownS&P 500, Cash, Oracle, AI consensus portfolioPractical references are shown beside the impossible hindsight ceiling when available.

Use this as benchmark evidence, not an investable strategy result. More resolved rounds are needed before making strong performance claims.

Fair model comparisons

Benchmark Comparison Sets

Sets are living groups. Older sets keep adding shared rounds, while newer model rosters become current automatically after enough shared results.

View all sets

Forming set Monthly Set: Sep 4, 2026

0 shared resolved rounds across 7 models.

3 more shared rounds to qualify Forming set Monthly Set: Sep 3, 2026

0 shared resolved rounds across 7 models.

3 more shared rounds to qualify Current benchmark Monthly Set: Aug 19, 2026

9 shared resolved rounds across 7 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: Aug 13, 2026

3 shared resolved rounds across 8 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: Jul 24, 2026

11 shared resolved rounds across 8 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: Jul 21, 2026

19 shared resolved rounds across 7 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: Jul 10, 2026

5 shared resolved rounds across 8 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: Jul 8, 2026

7 shared resolved rounds across 7 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: Jun 9, 2026

13 shared resolved rounds across 6 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: May 28, 2026

32 shared resolved rounds across 5 models.

Qualified at 3+ shared rounds Qualified set Monthly Set: May 10, 2026

35 shared resolved rounds across 4 models.

Qualified at 3+ shared rounds Current benchmark Weekly Set: Sep 4, 2026

7 shared resolved rounds across 7 models.

Qualified at 6+ shared rounds Forming set Weekly Set: Sep 3, 2026

1 shared resolved rounds across 7 models.

5 more shared rounds to qualify Qualified set Weekly Set: Aug 19, 2026

14 shared resolved rounds across 7 models.

Qualified at 6+ shared rounds Forming set Weekly Set: Aug 13, 2026

3 shared resolved rounds across 8 models.

3 more shared rounds to qualify Qualified set Weekly Set: Jul 24, 2026

11 shared resolved rounds across 8 models.

Qualified at 6+ shared rounds Qualified set Weekly Set: Jul 21, 2026

20 shared resolved rounds across 7 models.

Qualified at 6+ shared rounds Qualified set Weekly Set: Jul 10, 2026

6 shared resolved rounds across 8 models.

Qualified at 6+ shared rounds Qualified set Weekly Set: Jul 8, 2026

8 shared resolved rounds across 7 models.

Qualified at 6+ shared rounds Qualified set Weekly Set: Jun 9, 2026

15 shared resolved rounds across 6 models.

Qualified at 6+ shared rounds Qualified set Weekly Set: May 28, 2026

33 shared resolved rounds across 5 models.

Qualified at 6+ shared rounds Qualified set Weekly Set: May 24, 2026

35 shared resolved rounds across 4 models.

Qualified at 6+ shared rounds

Latest scored round

Most Recent Published Result

This chart shows the newest completed round only. Live rounds stay out of this chart until ending prices are collected.

Monthly result

Monthly Portfolio Returns

Models, S&P 500, and maximum possible return are shown on one scale.

Scored

Model portfolios S&P 500 benchmark Maximum possible return

GPT-5.6 Sol

Claude Fable 5

Grok 4.3

Claude Opus 5

Grok 4.6

Grok 4.5

Gemini 3.1 Pro

S&P 500

USO Crude Oil

GPT-5.6 Sol OpenAI

3.57%

Claude Fable 5 Anthropic

0.13%

Grok 4.3 xAI

-3.70%

Claude Opus 5 Anthropic

-4.80%

Grok 4.6 xAI

-6.35%

Grok 4.5 xAI

-6.41%

Gemini 3.1 Pro Google

-9.47%

S&P 500 Benchmark

0.69%

USO Crude Oil - Hindsight best asset

16.47%

Portfolio context

Shows each model's saved portfolio weights.

Model portfolios

Ranked in the same order as the chart.

GPT-5.6 Sol OpenAI

Semiconductors (SMH) 35% Cybersecurity (CIBR) 35% Real Estate (XLRE) 30%

Claude Fable 5 Anthropic

Semiconductors (SMH) 35% Regional Banks (KRE) 35% Industrials (XLI) 30%

Grok 4.3 xAI

US Dollar (UUP) 35% Small Value (IWN) 35% Utilities (XLU) 30%

Claude Opus 5 Anthropic

Industrials (XLI) 35% Regional Banks (KRE) 35% Small Value (IWN) 30%

Grok 4.6 xAI

Defense (ITA) 35% Utilities (XLU) 35% S&P 500 (SPY) 30%

Grok 4.5 xAI

Defense (ITA) 35% Regional Banks (KRE) 35% Industrials (XLI) 30%

Gemini 3.1 Pro Google

Solar (TAN) 35% Utilities (XLU) 35% Defense (ITA) 30%

Reference points

Not model portfolios.

S&P 500 Benchmark

Benchmark return over the same scoring window

USO Crude Oil - Hindsight best asset

100% Crude Oil (USO) hindsight ceiling

Run details

CB-2026-08-25-1M

2026-08-26 to 2026-09-25

Full result Benchmark results

WinnerGPT-5.6 Sol Return3.57% Models7 Eligible assets70

Benchmark result paths

Choose The Result View

Latest pages show one completed round. All-history pages are context. Comparison sets are the fair ranking view.

All rounds

Weekly result Latest Weekly

CB-2026-09-16-1W, 2026-09-17 to 2026-09-24

Scored Weekly aggregate Overall Weekly

71 all-available weekly rounds for context. Fair rankings use comparison sets.

7 rounds Monthly result Latest Monthly

CB-2026-08-25-1M, 2026-08-26 to 2026-09-25

Scored Monthly aggregate Overall Monthly

58 all-available monthly rounds for context. Fair rankings use comparison sets.

9 rounds

Live rounds

Waiting For Final Prices

These rounds are live or pending score. They are not counted in completed result charts yet.

Weekly live round

CB-2026-09-21-1W

Scores after the 2026-09-28 close.

2026-09-21 to 2026-09-28 official-v3-20260921-weekly

View locked portfolios

Monthly live round

CB-2026-09-21-1M

Scores after the 2026-10-21 close.

2026-09-21 to 2026-10-21 official-v3-20260921-monthly

View locked portfolios

Audit and data

Every Result Has An Audit Packet

Results link back to prompts, model outputs, portfolio decisions, prices, audit hashes, and scoring records.

Round index 146 public rounds with dates, prices, and audit packets. Scoring rules How model portfolios are compared with market returns. API docs Published data endpoints for results, rounds, models, and exposures. GitHub repository Source files, generated reports, and public benchmark artifacts.