A peer-normalized comparison of each model's typical allocation style across eligible official frozen portfolios, with current open positioning shown separately from historical behavior.
Every label, sentence, and pill comes from the same deterministic evidence record. Realized investment results are deliberately excluded from allocation-style classification.
Risk taking averaged 80.0/100, with a median 6.6 points above same-round peers; the difference had the same direction in 83% of 86 matched portfolios. Portfolios averaged 4.8 holdings, a 34.9% largest position, and 47.1% turnover.
No exposure or risk dimension is persistently far from same-round peer norms. Portfolios averaged 3.7 holdings, a 40.2% largest position, and 52.9% turnover.
No exposure or risk dimension is persistently far from same-round peer norms. Portfolios averaged 3.4 holdings, a 39.9% largest position, and 59.9% turnover.
Defensive assets averaged 16.6%, with a median 15.0 percentage points above same-round peers; the difference had the same direction in 76% of 70 matched portfolios. Portfolios averaged 4.9 holdings, a 31.8% largest position, and 50.1% turnover.
Risk taking averaged 71.0/100, with a median 6.4 points above same-round peers; the difference had the same direction in 74% of 27 matched portfolios. Portfolios averaged 4.1 holdings, a 36.5% largest position, and 70.6% turnover.
Real assets averaged 38.2%, with a median 10.0 percentage points above same-round peers; the difference had the same direction in 71% of 31 matched portfolios. Portfolios averaged 4.5 holdings, a 31.3% largest position, and 55.3% turnover.
Risk taking averaged 69.5/100, with a median 5.1 points below same-round peers; the difference had the same direction in 78% of 81 matched portfolios. Portfolios averaged 4.7 holdings, a 30.5% largest position, and 45.4% turnover.
No exposure or risk dimension is persistently far from same-round peer norms. Portfolios averaged 4.9 holdings, a 28.0% largest position, and 55.7% turnover.
Emerging pattern across 10 official portfolios. Portfolios averaged 4.8 holdings, a 32.5% largest position, and 40.6% turnover.
Only 10 peer-matched portfolios across 5 independent decision dates are available; stable labels require 8 and 6, respectively.These are cumulative allocation-behavior measures across eligible official saved portfolios. Performance remains available on the weekly and monthly leaderboards, but does not determine these behavior labels.
| Model | Risk | Holdings | Top holding | High risk | Defensive | Peer overlap | Turnover |
|---|---|---|---|---|---|---|---|
| GPT-5.5 | 80.0 / 100 | 4.83 | 34.9% | 84.5% | 4.1% | 55.0% | 47.1% |
| Grok 4.3 | 75.3 / 100 | 3.69 | 40.2% | 70.5% | 5.4% | 55.0% | 52.9% |
| Gemini 3.1 Pro | 72.1 / 100 | 3.43 | 39.9% | 65.3% | 11.7% | 48.6% | 59.9% |
| Claude Opus 4.7 | 71.3 / 100 | 4.90 | 31.8% | 63.6% | 16.6% | 60.6% | 50.1% |
| GPT-5.6 Sol | 71.0 / 100 | 4.07 | 36.5% | 74.4% | 7.4% | 48.3% | 70.6% |
| Grok 4.5 | 71.0 / 100 | 4.55 | 31.3% | 78.9% | 2.6% | 56.5% | 55.3% |
| Claude Opus 4.8 | 69.5 / 100 | 4.69 | 30.5% | 51.5% | 12.4% | 55.8% | 45.4% |
| Claude Fable 5 | 67.7 / 100 | 4.86 | 27.9% | 53.5% | 11.7% | 55.4% | 55.7% |
| Claude Opus 5 | 63.6 / 100 | 4.80 | 32.5% | 30.5% | 12.5% | 57.5% | 40.6% |
Each finding is tied to model IDs and metric keys in the generated report.
GPT-5.5 stands out by risk appetite at 80.0 / 100, while Gemini 3.1 Pro stands out by portfolio structure with a 39.9% average largest holding.
Claude Opus 5 has the highest defensive allocation at 12.5%. Claude Opus 5 has the lowest measured turnover at 40.6%.
Claude Opus 5 has the highest average peer overlap at 57.5%. This means its allocation weights have looked more like the rest of the roster than the most distinctive models.
The report is rebuilt from eligible official frozen portfolios during every publication build. No model receives a manually assigned caption, and the model's own descriptive wording cannot assign its label.
For each model and round, CapitalBench subtracts the median behavior-metric value of the other models in that same round. A behavior signal must exceed its published materiality floor, point in the same direction in at least 65% of matched portfolios, and have at least 8 matched portfolios across 6 independent decision dates.
Qualifying signals are ordered by absolute median peer difference divided by their materiality floor, then by persistence and a stable metric key. The strongest exposure or risk signal supplies the label modifier; peer-normalized construction, turnover, or overlap supplies the allocation-style noun.
Evidence is “established” only after 16 decision dates and 75% persistence. Opposite material weekly and monthly signals are marked horizon-dependent; a sufficiently sampled reversal under the newest methodology is marked evolving. The four pills always report signature, construction, tempo, and current open positioning (or lifecycle for a retired model). “Typical” uses all eligible history; “Now” uses only currently open portfolios.
Realized returns, ranks, ineligible or pilot runs, market-briefing prose, and free-form rationale wording are not classification inputs. Structured candidate-ledger, forecast, confidence, and key-risk fields are retained as decision-process context when coverage exists, but they do not override allocation evidence. Page-level “most” leader cards use active models only; retired profiles remain available as historical evidence.
Method version: capitalbench_behavior_evidence_v2
Peer baseline: leave-one-model-out same-round peer median
Wording provenance: deterministic_source_of_truth
Prompt contract: capitalbench_model_patterns_prompt_v2
A median same-round peer difference must meet the relevant floor before persistence can qualify it.
Average allocation-weighted risk appetite across all official saved portfolios. Higher means more growth, momentum, cyclical, and high-risk exposure.
Average number of non-zero assets in the model's official saved portfolios.
Average size of the largest single holding in each official saved portfolio.
Average allocation to assets rated as higher risk by the CapitalBench asset risk model.
Average allocation to cash, bonds, defensive sectors, and other lower-risk ballast.
Average allocation to technology, semiconductors, Nasdaq-style growth, and AI-linked technology exposure.
Average allocation to cash-like assets and duration-sensitive bond exposure.
Average allocation to non-U.S. country, regional, or international equity exposure.
Average allocation to commodities, crypto, energy, gold, and other inflation-linked or real-asset groups.
Average allocation to the S&P 500 benchmark option across official saved portfolios.
Average cosine similarity between this model's allocation weights and peer model portfolios in the same rounds.
Average one-half summed absolute allocation change between consecutive same-track portfolios.
Average finishing rank across resolved rounds. Lower is better.
Average model score versus the hindsight-best eligible asset in each resolved round.
