Anthropic
Anthropic

Claude Opus 4.8

anthropic-claude-opus-4-8

Retired from new benchmark rounds. Historical portfolios and scores remain part of the public record.

Portfolio pattern Risk-conscious steady allocator Risk taking averaged 69.2/100, with a median 4.4 points below same-round peers; the difference had the same direction in 74% of 99 matched portfolios. Portfolios averaged 4.3 holdings, a 35.3% largest position, and 43.4% turnover.
Retired from new rounds WeeklyMonthly First run 2026-05-28
Live portfolios 9 9 open rounds
Completed rounds 90 scored rounds
Beat S&P 500 38.9% all completed tracks
Portfolio Minus S&P 500 -0.57 pp completed average
Portfolio pattern

How Does This Model Tend To Invest?

Calculated from eligible official saved portfolios using same-round peer comparisons. How labels and pills are determined

Saved portfolios 99
Average holdings 4.3
Average largest position 35.3%
Most common top holding Semiconductors (SMH)
Typical approach Risk-conscious steady allocator

Risk taking averaged 69.2/100, with a median 4.4 points below same-round peers; the difference had the same direction in 74% of 99 matched portfolios. Portfolios averaged 4.3 holdings, a 35.3% largest position, and 43.4% turnover.

Moderate evidence · based on 99 saved portfolios.
Risk 69/100 · −4 pts vs peers4.3 holdings · 35% top43% turnoverHistorical · retired
Risk taking 69.2 / 100 0 defensive to 100 aggressive
Typical largest position 35.3% average top holding
Typical holdings 4.3 average non-zero positions
Portfolio turnover 43.4% average change between rounds
Portfolio Difference / Combined

How different is this portfolio from the group?

A score of 77.8 means about 77.8% of allocation would need to change to match the average portfolio selected by the other models.

Compare every model
77.8 / 100
0 same as group to 100 completely different
monthly 65.0 2 same-round comparisons
weekly 90.7 2 same-round comparisons
Recent-winner tilt · Combined

Does this model follow recent winners?

Mixed recent-performance exposure. Combined gives monthly and weekly behavior equal weight. The score uses only prices available before each portfolio was frozen.

50.0 / 100
+31.1 points vs peers
monthly 50.0 0.00% in the top 20%
weekly 50.0 0.00% in the top 20%
Structured decision process

How the model builds its shortlist

Candidate-ledger evidence is available for 33 of 99 eligible portfolios. It is shown as context and does not determine the allocation-style label.

Avg candidates8.0 SPY included100.0% Forecast spread8.0% Stated confidence55.5%
Coverage — forecasts 33/99 · confidence 99/99 · key risks 99/99
Overall risk appetite

Growth

Usually favors equities, sectors, or thematic growth exposure.

3.60 / 5 allocation-weighted score
Claude Opus 4.83.60
46.5% high-risk exposure 18.3% technology exposure 11.7% defensive holdings
Most frequently held assets
S&P 500 (SPY)US Broad Market 54.5% held 18.9% avg
Healthcare Sector (XLV)US Sector 51.5% held 12.9% avg
Financials Sector (XLF)US Sector 47.5% held 11.4% avg
Semiconductors (SMH)AI and Technology 32.3% held 9.1% avg
Industrials Sector (XLI)US Sector 28.3% held 5.5% avg
Energy Sector (XLE)US Sector 18.2% held 4.5% avg
US Dividend Equities (SCHD)US Factor Equity 18.2% held 3.8% avg
Average allocation by category
US Sector39.8%
US Broad Market20.9%
AI and Technology11.1%
US Factor Equity9.3%
US Industry3.2%
Country Equity3.2%
US Style Factor2.9%
Current portfolios

What Is This Model Holding Now?

9 open portfolios across weekly and monthly tests. Completed rounds are excluded.

Monthly9 open portfolios
11 assets
Completed rounds are excluded from this live view.Next scoring target: 2026-08-31
Still in progress

How Are Its Open Portfolios Doing?

Latest available close for live rounds only. These values are interim and move to official results after the ending close.

Completed results

How Has This Model Performed?

Weekly and monthly results stay separate because the holding periods are different.

Weekly record

9 wins / 50 completed

Avg return -0.12%
S&P 500 +0.21%
Avg Portfolio Minus S&P 500 -0.33 pp
Hit rate 40.0%
Avg rank 3.8
Best round CB-2026-06-13-1W
Monthly record

10 wins / 40 completed

Avg return +0.61%
S&P 500 +1.48%
Avg Portfolio Minus S&P 500 -0.87 pp
Hit rate 37.5%
Avg rank 3.0
Best round CB-2026-07-21-1M
Round by round

When Did It Beat The S&P 500?

Bars to the right beat the S&P 500. Bars to the left trailed it.

CB-2026-05-28-1Mmonthly / rank 2 of 5 +3.25 pp CB-2026-05-28-1Wweekly / rank 2 of 5 +3.25 pp CB-2026-06-01-1Mmonthly / rank 1 of 5 +3.16 pp CB-2026-05-29-1Mmonthly / rank 1 of 5 +1.68 pp CB-2026-06-01-1Wweekly / rank 1 of 5 -1.88 pp CB-2026-05-29-1Wweekly / rank 1 of 5 -2.06 pp CB-2026-06-02-1Mmonthly / rank 4 of 5 -0.98 pp CB-2026-06-02-1Wweekly / rank 1 of 5 -0.72 pp CB-2026-06-03-1Mmonthly / rank 2 of 5 -3.87 pp CB-2026-06-03-1Wweekly / rank 1 of 5 -4.21 pp CB-2026-06-05-1Mmonthly / rank 4 of 5 -0.51 pp CB-2026-06-05-1Wweekly / rank 2 of 5 +0.29 pp CB-2026-06-08-1Mmonthly / rank 2 of 5 +1.70 pp CB-2026-06-08-1Wweekly / rank 2 of 5 -1.96 pp CB-2026-06-09-1Mmonthly / rank 1 of 6 +0.85 pp CB-2026-06-09-1Wweekly / rank 3 of 6 -1.47 pp CB-2026-06-12-1Mmonthly / rank 2 of 6 -1.13 pp CB-2026-06-12-1Wweekly / rank 5 of 6 -2.46 pp CB-2026-06-13-1Mmonthly / rank 3 of 5 -3.60 pp CB-2026-06-13-1Wweekly / rank 2 of 6 +3.87 pp CB-2026-06-15-1Mmonthly / rank 3 of 5 -6.14 pp CB-2026-06-15-1Wweekly / rank 5 of 5 +3.25 pp CB-2026-06-16-1Mmonthly / rank 3 of 5 -8.61 pp CB-2026-06-16-1Wweekly / rank 3 of 5 +0.78 pp CB-2026-06-17-1Mmonthly / rank 1 of 5 -6.44 pp CB-2026-06-17-1Wweekly / rank 2 of 5 -0.18 pp CB-2026-06-18-1Mmonthly / rank 1 of 5 -5.02 pp CB-2026-06-18-1Wweekly / rank 1 of 5 -0.50 pp CB-2026-06-22-1Mmonthly / rank 3 of 5 -6.80 pp CB-2026-06-22-1Wweekly / rank 1 of 5 -1.01 pp CB-2026-06-23-1Mmonthly / rank 5 of 5 -2.11 pp CB-2026-06-23-1Wweekly / rank 5 of 5 +1.62 pp CB-2026-06-24-1Mmonthly / rank 1 of 5 -0.13 pp CB-2026-06-24-1Wweekly / rank 1 of 5 +0.57 pp CB-2026-06-25-1Mmonthly / rank 2 of 5 -1.20 pp CB-2026-06-25-1Wweekly / rank 3 of 5 +1.34 pp CB-2026-06-26-1Mmonthly / rank 1 of 5 +0.82 pp CB-2026-06-26-1Wweekly / rank 5 of 5 -1.33 pp CB-2026-06-29-1Mmonthly / rank 2 of 5 -2.61 pp CB-2026-06-29-1Wweekly / rank 1 of 5 -1.12 pp CB-2026-06-30-1Mmonthly / rank 1 of 5 +2.09 pp CB-2026-06-30-1Wweekly / rank 2 of 5 -3.27 pp CB-2026-07-01-1Mmonthly / rank 2 of 6 +0.94 pp CB-2026-07-01-1Wweekly / rank 5 of 6 +0.45 pp CB-2026-07-02-1Mmonthly / rank 3 of 6 +0.05 pp CB-2026-07-02-1Wweekly / rank 4 of 6 -1.60 pp CB-2026-07-06-1Mmonthly / rank 1 of 6 -0.92 pp CB-2026-07-06-1Wweekly / rank 3 of 6 -1.79 pp CB-2026-07-07-1Mmonthly / rank 5 of 6 -1.46 pp CB-2026-07-07-1Wweekly / rank 5 of 6 -1.76 pp CB-2026-07-08-1Mmonthly / rank 5 of 7 -1.77 pp CB-2026-07-08-1Wweekly / rank 5 of 7 -0.87 pp CB-2026-07-09-1Mmonthly / rank 3 of 7 -1.19 pp CB-2026-07-09-1Wweekly / rank 1 of 7 -2.00 pp CB-2026-07-10-1Mmonthly / rank 1 of 8 -2.82 pp CB-2026-07-10-1Wweekly / rank 2 of 8 -3.05 pp CB-2026-07-13-1Mmonthly / rank 8 of 8 -0.25 pp CB-2026-07-13-1Wweekly / rank 8 of 8 +0.99 pp CB-2026-07-14-1Mmonthly / rank 3 of 8 +1.48 pp CB-2026-07-14-1Wweekly / rank 5 of 8 +2.23 pp CB-2026-07-15-1Mmonthly / rank 7 of 8 +2.38 pp CB-2026-07-15-1Wweekly / rank 7 of 8 +2.77 pp CB-2026-07-17-1Mmonthly / rank 3 of 8 -0.42 pp CB-2026-07-17-1Wweekly / rank 3 of 8 +1.76 pp CB-2026-07-20-1Wweekly / rank 4 of 8 +0.86 pp CB-2026-07-21-1Mmonthly / rank 3 of 7 +3.39 pp CB-2026-07-21-1Wweekly / rank 3 of 7 +0.91 pp CB-2026-07-22-1Mmonthly / rank 3 of 7 +3.22 pp CB-2026-07-22-1Wweekly / rank 4 of 7 +2.01 pp CB-2026-07-23-1Mmonthly / rank 4 of 7 +2.04 pp CB-2026-07-23-1Wweekly / rank 6 of 7 -1.00 pp CB-2026-07-24-1Mmonthly / rank 4 of 8 +1.42 pp CB-2026-07-24-1Wweekly / rank 5 of 8 -0.92 pp CB-2026-07-27-1Mmonthly / rank 6 of 8 -0.64 pp CB-2026-07-27-1Wweekly / rank 6 of 8 -2.25 pp CB-2026-07-28-1Mmonthly / rank 3 of 8 -1.68 pp CB-2026-07-28-1Wweekly / rank 7 of 8 -4.46 pp CB-2026-07-29-1Mmonthly / rank 3 of 8 -2.06 pp CB-2026-07-29-1Wweekly / rank 3 of 8 -4.61 pp CB-2026-07-30-1Mmonthly / rank 6 of 8 -0.67 pp CB-2026-07-30-1Wweekly / rank 2 of 8 -1.23 pp CB-2026-07-31-1Wweekly / rank 2 of 8 -0.86 pp CB-2026-08-04-1Wweekly / rank 8 of 8 -0.23 pp CB-2026-08-05-1Wweekly / rank 6 of 8 +0.33 pp CB-2026-08-07-1Wweekly / rank 7 of 8 +0.21 pp CB-2026-08-09-1Wweekly / rank 7 of 8 -0.00 pp CB-2026-08-11-1Wweekly / rank 6 of 8 +1.37 pp CB-2026-08-13-1Wweekly / rank 5 of 8 +3.34 pp CB-2026-08-15-1Wweekly / rank 7 of 8 0.00 pp CB-2026-08-18-1Wweekly / rank 5 of 8 0.00 pp
Ranking context

Where Does It Rank Against Comparable Models?

Each group uses only rounds completed by every included model.

See every comparison group 14
Weekly · Current

Jul 24, 2026 roster

Weekly comparison set automatically opened when the Jul 24 official roster first required a new equal-run benchmark group across 8 models.

Open full set
#7 Rank in set 6.9 CapitalBench Score 7.9% Total return 11 Shared rounds

Current Weekly Benchmark

CapitalBench Score

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

GPT-5.6 Sol
Grok 4.5
Grok 4.3
Claude Opus 5
GPT-5.5
Claude Fable 5
Claude Opus 4.8 This model
Gemini 3.1 Pro
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

GPT-5.6 Sol OpenAI · 11/11 scored rounds
13.8
Grok 4.5 xAI · 11/11 scored rounds
12.8
Grok 4.3 xAI · 11/11 scored rounds
11.4
Claude Opus 5 Anthropic · 11/11 scored rounds
9.6
GPT-5.5 OpenAI · 11/11 scored rounds
8.8
Claude Fable 5 Anthropic · 11/11 scored rounds
8.4
Claude Opus 4.8 Anthropic · 11/11 scored rounds
6.9
Gemini 3.1 Pro Google · 11/11 scored rounds
2.9
S&P 500 S&P 500 · 11/11 scored rounds
17.8
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
11 shared resolved rounds8 equal-run models rankedQualified at 6+ shared roundsNewest included round: CB-2026-08-11-1W
Return context

Average Return Details

Average portfolio return across the same finished rounds.

OpenAI GPT-5.6 Sol
1.45%
xAI Grok 4.5
1.35%
xAI Grok 4.3
1.20%
Anthropic Claude Opus 5
1.01%
OpenAI GPT-5.5
0.92%
Anthropic Claude Fable 5
0.88%
Anthropic Claude Opus 4.8
0.72%
Google Gemini 3.1 Pro
0.30%
S&P S&P 500
1.87%
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.
10.52%
Excluded for fairness: CB-2026-08-13-1W missing openai-gpt-5-5; CB-2026-08-15-1W missing openai-gpt-5-5; CB-2026-08-18-1W missing openai-gpt-5-5; CB-2026-08-19-1W missing anthropic-claude-opus-4-8, openai-gpt-5-5; CB-2026-08-20-1W missing anthropic-claude-opus-4-8, openai-gpt-5-5; CB-2026-08-21-1W missing anthropic-claude-opus-4-8, openai-gpt-5-5 Fairness rule: every ranked model completed every included round. A missed round is excluded from this set for everyone.
Model-specific findings

What Has CapitalBench Learned About Claude Opus 4.8?

Findings tied directly to this model's decisions or completed results.

Model BehaviorAs of Aug 30
Model behavior profiles10 models

Model allocation styles are separating into clear behavior profiles

GPT-5.5 has the highest average risk-taking score at 79.7/100. Grok 4.6 has the largest average top holding at +69.79%. Claude Opus 4.8 has the lowest measured turnover at +43.35%.

Momentum exposure measures how much of the frozen portfolio went into assets that had already been recent winners before the model made its allocation.

High confidenceMath: deterministicData through Aug 30, 2026
Highest Average Risk Taking Score
79.7/100
Largest Average Top Holding
+69.8%
Lowest Average Turnover
+43.4%
Market EnvironmentAs of Aug 28
Monthly environment consistency22 resolved rounds1 model

Claude Opus 4.8 has the strongest monthly score floor

Its lowest CapitalBench Score across 3 tested market directions is -14.2, with at least 5 model observations in each included direction.

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 Aug 28, 2026
Score Floor
-14.2
Average Model Return
-0.39%
Directions Covered
3
Market EnvironmentAs of Aug 28
Weekly up/down split10 resolved rounds1 model

Claude Opus 4.8 changes most between weekly up and down environments

The model averaged -0.40% in down environments and +1.01% in up environments, a 1.4 percentage-point gap.

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 Aug 28, 2026
Average Return Gap
+1.41%
Down Average Return
-0.40%
Up Average Return
+1.01%
Decision history

What Did It Choose In Each Round?

Open a row to see the full portfolio, rationale, and audit files.

Round Track Window Portfolio Result
CB-2026-08-18-1W official-v3-20260818-weekly weekly 2026-08-18 to 2026-08-25 SP500 100.0% 0.00 pp
Full portfolio
S&P 500 (SPY) 100.0%
Rationale

SPY is essentially flat over 5 sessions but up ~4% over 21, with broad positive breadth. Macro data are mixed: soft payrolls (-23k), weak retail sales and sentiment, but firm ISM and elevated inflation with a hawkish-leaning Fed. Dispersion is moderate. No strong basis for a single-name overreaction bet that reliably beats SPY over one week; most slate reversal candidates lack independent in-window support.

CB-2026-08-18-1M official-v3-20260818-monthly-clean monthly 2026-08-18 to 2026-09-18 SP500 100.0% Pending
Full portfolio
S&P 500 (SPY) 100.0%
Rationale

Macro is soft-landing-with-friction: weak payrolls (-23k, downward revisions), cooling core PCE 3.3% but sticky, hawkish Fed dividing 9-3, and consumer sentiment falling to 51.0. SPY up 13% YTD near highs. Dispersion is high with tech/semis, precious metals momentum, and defensive rate-sensitives all diverging. No compelling single-name overreaction warrants overriding SPY; edges are thin and mostly no_edge, so SPY default is preferred.

CB-2026-08-15-1W official-v3-20260815-weekly-clean weekly 2026-08-17 to 2026-08-24 SP500 100.0% 0.00 pp
Full portfolio
S&P 500 (SPY) 100.0%
Rationale

SPY near all-time highs with modest weekly gains; breadth positive (RSP slightly outperforming). Macro mixed: soft labor data, cooling core inflation, but yields drifting up. No high-conviction overreaction with strong reversal setup among the slate. Retailer earnings (HD, WMT, TGT, LOW) and FOMC minutes create idiosyncratic risk. Defaulting toward SPY given weak edge.

CB-2026-08-15-1M official-v3-20260815-monthly-clean monthly 2026-08-17 to 2026-09-17 SP500 100.0% Pending
Full portfolio
S&P 500 (SPY) 100.0%
Rationale

SPY near all-time highs after +13.7% YTD, with cooling labor market (-23k payrolls, downward revisions), sticky inflation (CPI 3.4%, core PCE 3.3%), and a hawkish-hold FOMC. High cross-sectional dispersion with energy/commodities/foreign strong and rate-sensitive defensives weak. The monthly window includes CPI, PPI, jobs, and a Sept 16 FOMC decision, all high-variance catalysts. Few candidates show durable, supported edge over SPY; most extreme moves lack independent in-window support, so a SPY-heavy stance is warranted.

CB-2026-08-13-1W official-v2-2-all-weekly-20260813 weekly 2026-08-13 to 2026-08-20 HEALTHCARE 35.0%SP500 35.0%GOLD 30.0% +3.34 pp
Full portfolio
Healthcare Sector (XLV) 35.0% S&P 500 (SPY) 35.0% Gold (IAU) 30.0%
Rationale

Healthcare and gold both show strong recent active returns with low SPY correlation, providing diversification while SPY anchors the core.

CB-2026-08-13-1M official-v2-2-all-monthly-20260813 monthly 2026-08-13 to 2026-09-14 SP500 40.0%HEALTHCARE 25.0%GOLD 20.0%DIVIDEND 15.0% Pending
Full portfolio
S&P 500 (SPY) 40.0% Healthcare Sector (XLV) 25.0% Gold (IAU) 20.0% US Dividend Equities (SCHD) 15.0%
Rationale

Core SPY with defensive quality and gold overweight to capture recent leadership while managing macro risk.

CB-2026-08-11-1W official-v2-2-all-weekly-20260811 weekly 2026-08-11 to 2026-08-18 HEALTHCARE 30.0%GOLD 25.0%SP500 25.0%ENERGY 20.0% +1.37 pp
Full portfolio
Healthcare Sector (XLV) 30.0% Gold (IAU) 25.0% S&P 500 (SPY) 25.0% Energy Sector (XLE) 20.0%
Rationale

Tilt toward defensives and real assets given elevated inflation prints scheduled and geopolitical energy risk, while retaining SPY core.

CB-2026-08-11-1M official-v2-2-all-monthly-20260811 monthly 2026-08-11 to 2026-09-11 SP500 40.0%HEALTHCARE 25.0%FINANCIALS 20.0%DIVIDEND 15.0% Pending
Full portfolio
S&P 500 (SPY) 40.0% Healthcare Sector (XLV) 25.0% Financials Sector (XLF) 20.0% US Dividend Equities (SCHD) 15.0%
Rationale

Overweight defensive-quality equity sectors with strong relative trends and low drawdown while keeping SPY as core, aiming for modest alpha with reduced volatility.

CB-2026-08-09-1W official-v2-2-all-weekly-20260809 weekly 2026-08-10 to 2026-08-17 SP500 50.0%GOLD 25.0%HEALTHCARE 25.0% -0.00 pp
Full portfolio
S&P 500 (SPY) 50.0% Gold (IAU) 25.0% Healthcare Sector (XLV) 25.0%
Rationale

Half SPY, quarter gold for momentum/safe-haven, quarter healthcare for defensive tilt given soft payrolls.

CB-2026-08-09-1M official-v2-2-all-monthly-20260809-clean monthly 2026-08-10 to 2026-09-10 SP500 45.0%SOFTWARE 30.0%BIOTECH 25.0% Pending
Full portfolio
S&P 500 (SPY) 45.0% Software (IGV) 30.0% Biotechnology (XBI) 25.0%
Rationale

Overweight SPY plus biotech and software tilts that clear the SPY base hurdle, exploiting strong quality-evidence scores while respecting cluster caps.

Audit and data Public Audit Packets For This Model