TypeSafe Jev 1.13
A decision model that assigns probabilities to named outcomes using supplied context. Jev complements numerical forecasters with a different kind of answer; it does not generate a sequence of future values.
typesafe/jev-1.13Different models bring different learned patterns. Compare numerical forecasts and, separately, Jev’s estimates for named outcomes.
“Analyst” is an analogy for a model opinion. The cards separate what a model learned from evidence about where it performs well.
A decision model that assigns probabilities to named outcomes using supplied context. Jev complements numerical forecasters with a different kind of answer; it does not generate a sequence of future values.
typesafe/jev-1.13A financial-market specialist trained on open, high, low, close and trading activity, rather than a general-purpose numerical model.
NeoQuasar/Kronos-baseA general numerical forecaster with joint multivariate and covariate support. Its broad training gives it exposure to different kinds of temporal patterns; it is not a dedicated stock, crypto or options analyst.
amazon/chronos-2A general forecasting perspective built around numerical sequences, rather than financial news or company fundamentals. We do not infer financial specialization from the scale of its pretraining.
google/timesfm-2.5A recurrent forecasting model that brings another approach to patterns within and across numerical series. Broad pretraining is a background in time-series behavior, not evidence of specialist expertise in any particular asset.
nxai/tirex-2At each future step, Pretick takes the median of the participating point predictions. With two models, this is their average. It is a transparent starting point, not a learned weighting or a language-model arbiter. The response includes every component opinion and any exclusions. At least two models must succeed.
TiRex-2 is excluded above 320 future points; TimesFM is excluded from requests with past covariates. Read the API limits or the machine-readable catalog. Jev uses a separate outcome endpoint and is not included in numerical consensus. Each model card describes its access and licensing.