Multi model forecasts
Independent probability estimates from four production models (Claude, GPT, Gemini, Grok) for binary forecast objects. No model sees another model output.
Inspect the public forecast corpus, model probabilities and resolved outcomes behind Crene’s living decision records.
Independent forecasts. Named resolution sources. Preserved histories.
Fetching the latest published record.
The broad corpus is not the leakage controlled benchmark or the macro only subset. Scored binary outcomes and structural records remain separate.
Total resolved questions refers to the full Crene native resolved set. Model sample sizes count each model’s scored rows, so they may differ from the consensus sample.
The directional accuracy numerator and denominator will appear when the public statistics are available.
View methodology and benchmark definitions ↗Performance on resolved binary events, measured using Brier score and directional accuracy. Lower Brier is better.
Loading the public record…
Events provide scored outcomes. Thesis, Factor and Scenario records organize the assumptions, distributions and pathways around them.
Loading the public record…
Independent probability estimates from four production models (Claude, GPT, Gemini, Grok) for binary forecast objects. No model sees another model output.
Binary outcome questions across macro releases (CPI, NFP, PMI), central bank decisions, commodities, crypto, policy, and market events. Each resolved against a named source.
One live Warsh map and two frozen AI records decompose yes or no investment views into assumptions. Each record distinguishes current model reads from immutable weekly freezes.
Three active continuous distributions track the S&P 500, US 10Y Treasury yield, and US GDP attributable to AI, with recent movement, uncertainty, model disagreement, and driver review context.
Four archived scenario records preserve 400 authored pathways across AI labor, US primacy, European rearmament, and India growth. Pathway counts describe authored coverage, not probability.
Consensus probability is snapshotted daily for binary forecast objects. The full evolution of the forecast is preserved through resolution.
Each model is scored on resolved binary events. Per model performance is measured continuously, not curated. Continuous factors and long-horizon scenario structures require different calibration metrics as outcomes accrue.
New investment thesis maps, factors, and scenario components expand the corpus alongside macro, policy, market, and AI-related event flow. Resolutions compound the calibration signal.
Many questions can still depend on a small set of shared drivers. Question density shows how tightly each record’s components cluster by domain.
Question density is a structural read, not a correlation estimate. Exact effective independence and covariance figures are not published before the observation record can support them.
Loading the public record…
Every outcome is tied to a verifiable source. Binary events resolve against a tiered source allowlist, with the source for each outcome named and auditable.
Government statistical agencies, central banks, regulators. SEC filings, BLS releases, Fed statements, BEA, BoJ, ECB.
Major news wires reporting primary releases. Reuters, Bloomberg, AP, agency wire confirmations.
Used only to confirm Tier A or B. No event resolves on a single Tier C source.
Structural representations remain separate from scored evidence until outcomes accrue. Continuous factors and long horizon scenarios require different calibration metrics.
Explore the sample exports or request data for an internal research workflow.
Access includes the resolved event archive, per event model breakdown, calibration history and thesis record structure behind the private review workflow.
Crene is listed on Neudata, Eagle Alpha and Monda for institutional procurement.
Review a thesis with Crene ↗