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

36 completed 25 live

Weekly and monthly tracks are scored separately.

Current weekly benchmark leader

Claude Opus 4.8 -4.5 CapitalBench Score

Latest scored CB-2026-07-02-1W Latest live CB-2026-07-10-1W / CB-2026-07-10-1M Models 8 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 Jul 9

Weekly market environments12 resolved rounds2 modelsReady sample

Weekly model leadership changes with the S&P 500 environment

Claude Opus 4.8 leads down environments at -2.70% across 8 tests; GPT-5.5 leads up environments at +2.11% across 4 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 Jul 9, 2026

Down Leader Average Return: -2.70%
Down Shared Rounds: 8
Down Leader Stability: 0.88

Confidence CalibrationAs of Jul 10

All resolved official results36 resolved rounds181 scored resultsMedian confidence 0.55Resolved history

High-confidence model calls have underperformed lower-confidence calls

Across resolved official results, submissions at or above the median confidence of 0.55 averaged -0.57%, while lower-confidence submissions averaged -0.36%.

Confidence is the model's own 0-1 self-reported confidence at submission time, compared with later realized returns.

High confidenceMath: deterministicData through Jul 10, 2026

High Confidence Average Return: -0.57%
Low Confidence Average Return: -0.36%
High Confidence Average Capitalbench Score: -15.7

Consensus PerformanceJul 2-Jul 9

Weekly resultCB-2026-07-02-1W6 modelsOracle: Crude Oil (USO), +4.84%Resolved result

AI consensus portfolio scored -10.8 versus the oracle

If the weekly model allocations were averaged into one consensus portfolio, it returned -0.52% versus +0.93% for the S&P 500 and +4.84% 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 Jul 9, 2026

Consensus Portfolio Return: -0.52%
Average Model Return: -0.52%
Consensus Capitalbench Score: -10.8

All insights

Weekly track

Claude Opus 4.8 Leads

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

23 scored

Claude Opus 4.8 Anthropic

CapitalBench Score -4.5 Avg return leader -0.38% Shared rounds 23 Timeline One market week

Latest result Comparison sets All history Audit packet

Monthly track

Claude Opus 4.8 Leads

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

8 scored

Claude Opus 4.8 Anthropic

CapitalBench Score 0.5 Avg return leader 0.11% Shared rounds 8 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 23 weekly rounds.

23 shared resolved rounds5 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-07-02-1W

Shared resolved rounds

CapitalBench Score

Every ranked model has the same included rounds. Calculation.

Claude Opus 4.8

Grok 4.3

Claude Opus 4.7

GPT-5.5

Gemini 3.1 Pro

S&P 500

Max possible What is this? hindsight best asset

Higher benchmark score is better.

Claude Opus 4.8 Anthropic · 23/23 scored rounds

-4.5

Grok 4.3 xAI · 23/23 scored rounds

-7.0

Claude Opus 4.7 Anthropic · 23/23 scored rounds

-8.4

GPT-5.5 OpenAI · 23/23 scored rounds

-17.3

Gemini 3.1 Pro Google · 23/23 scored rounds

-19.7

S&P 500 S&P 500 · 23/23 scored rounds

-0.2

Max possible Hindsight ceiling, not a model portfolio

What is this? 100.0

23 shared resolved rounds5 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-07-02-1W

Return context

Average Return Details

Average portfolio return across the same finished rounds.

Claude Opus 4.8

-0.38%

Grok 4.3

-0.58%

Claude Opus 4.7

-0.70%

GPT-5.5

-1.45%

Gemini 3.1 Pro

-1.64%

S&P S&P 500

-0.02%

MAX Max possible What is this?

8.33%

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 8 monthly rounds.

8 shared resolved rounds5 equal-run models rankedQualified at 3+ shared roundsNewest included round: CB-2026-06-09-1M

Shared resolved rounds

CapitalBench Score

Every ranked model has the same included rounds. Calculation.

Claude Opus 4.8

Grok 4.3

Claude Opus 4.7

Gemini 3.1 Pro

GPT-5.5

S&P 500

Max possible What is this? hindsight best asset

Higher benchmark score is better.

Claude Opus 4.8 Anthropic · 8/8 scored rounds

0.5

Grok 4.3 xAI · 8/8 scored rounds

0.2

Claude Opus 4.7 Anthropic · 8/8 scored rounds

-0.4

Gemini 3.1 Pro Google · 8/8 scored rounds

-9.9

GPT-5.5 OpenAI · 8/8 scored rounds

-15.6

S&P 500 S&P 500 · 8/8 scored rounds

-2.6

Max possible Hindsight ceiling, not a model portfolio

What is this? 100.0

8 shared resolved rounds5 equal-run models rankedQualified at 3+ shared roundsNewest included round: CB-2026-06-09-1M

Return context

Average Return Details

Average portfolio return across the same finished rounds.

Claude Opus 4.8

0.11%

Grok 4.3

0.03%

Claude Opus 4.7

-0.08%

Gemini 3.1 Pro

-2.07%

GPT-5.5

-3.25%

S&P S&P 500

-0.55%

MAX Max possible What is this?

20.81%

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 evidence25 resolved rounds / 128 model resultsCurrent threshold met at 6+ rounds

Equal-run comparison5 models on the same 23 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 evidence11 resolved rounds / 53 model resultsCurrent threshold met at 3+ rounds

Equal-run comparison5 models on the same 8 roundsRanked models are compared only on rounds every model in the roster completed.

ProtocolMixed protocolCompleted history includes 10 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: Jul 10, 2026

0 shared resolved rounds across 8 models.

3 more shared rounds to qualify Forming set Monthly Set: Jul 8, 2026

0 shared resolved rounds across 7 models.

3 more shared rounds to qualify Forming set Monthly Set: Jun 9, 2026

1 shared resolved rounds across 6 models.

2 more shared rounds to qualify Current benchmark Monthly Set: May 28, 2026

8 shared resolved rounds across 5 models.

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

11 shared resolved rounds across 4 models.

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

0 shared resolved rounds across 8 models.

6 more shared rounds to qualify Forming set Weekly Set: Jul 8, 2026

0 shared resolved rounds across 7 models.

6 more shared rounds to qualify Forming set Weekly Set: Jun 9, 2026

5 shared resolved rounds across 6 models.

1 more shared rounds to qualify Current benchmark Weekly Set: May 28, 2026

23 shared resolved rounds across 5 models.

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

25 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.

Weekly result

Weekly 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

Gemini 3.1 Pro

Claude Fable 5

GPT-5.5

Claude Opus 4.8

Claude Opus 4.7

Grok 4.3

S&P 500

Max

Gemini 3.1 Pro Google

0.16%

Claude Fable 5 Anthropic

-0.14%

GPT-5.5 OpenAI

-0.23%

Claude Opus 4.8 Anthropic

-0.67%

Claude Opus 4.7 Anthropic

-0.74%

Grok 4.3 xAI

-1.50%

S&P 500 Benchmark

0.93%

Max possible USO

4.84%

Portfolio context

Shows each model's saved portfolio weights.

Model portfolios

Ranked in the same order as the chart.

Gemini 3.1 Pro Google

S&P 500 (SPY) 40% Healthcare (XLV) 20% Financials (XLF) 20% T-Bills (BIL) 20%

Claude Fable 5 Anthropic

Healthcare (XLV) 30% Financials (XLF) 20% Low Vol (SPLV) 20% Biotechnology (XBI) 15% S&P 500 (SPY) 15%

GPT-5.5 OpenAI

Biotechnology (XBI) 30% Healthcare (XLV) 25% Financials (XLF) 20% Defense (ITA) 15% Low Vol (SPLV) 10%

Claude Opus 4.8 Anthropic

Healthcare (XLV) 30% Financials (XLF) 25% Low Vol (SPLV) 20% Value (IWD) 15% T-Bills (BIL) 10%

Claude Opus 4.7 Anthropic

Healthcare (XLV) 30% Financials (XLF) 25% Low Vol (SPLV) 20% Equal-Weight S&P 500 (RSP) 15% T-Bills (BIL) 10%

Grok 4.3 xAI

Healthcare (XLV) 40% Staples (XLP) 30% Low Vol (SPLV) 30%

Reference points

Not model portfolios.

S&P 500 Benchmark

Benchmark return over the same scoring window

Max possible USO

100% Crude Oil (USO) hindsight ceiling

Run details

CB-2026-07-02-1W

2026-07-02 to 2026-07-09

Full result Benchmark results

WinnerGemini 3.1 Pro Return0.16% Models6 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-2026-07-02-1W, 2026-07-02 to 2026-07-09

Scored Weekly aggregate Overall Weekly

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

23 rounds Monthly result Latest Monthly

CB-2026-06-09-1M, 2026-06-09 to 2026-07-09

Scored Monthly aggregate Overall Monthly

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

8 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-2026-07-10-1W

Scores after the 2026-07-17 close.

2026-07-10 to 2026-07-17 official-20260710

View locked portfolios

Monthly live round

CB-2026-07-10-1M

Scores after the 2026-08-10 close.

2026-07-10 to 2026-08-10 official-20260710

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 61 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.