Predictalot

Install

Pick your client. Read the repository before you give any tool access to your data.

About

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this server can do and what it asks for, from its tool list, install config and code. Checked Oct 25, 2026.

Tools

6 read · 0 write · 0 delete · 1 not yet classified

Secrets

None asked for

Install scripts

None found

Runs as

Docker image

Tools (7)

Read from the source code; the running server may differ. Descriptions are the server’s own words; hidden characters are shown as [U+…].

• forecast_covariates_both_ensemble

  Past + future covariates ensemble. Args as forecast_covariates_both_<model> + per-slug weights. Currently only chronos-2 supports this type.

• forecast_covariates_future_ensemble

  Future-only covariates ensemble. Args as forecast_covariates_future_<model> + per-slug weights. Only useful once more than one backend supports the type.

• forecast_covariates_past_ensemble

  Past-only covariates ensemble. Args as forecast_covariates_past_<model> + per-slug weights.

• forecast_multivariate_ensemble

  Multivariate ensemble — weighted mean across every model that supports multivariate forecasting. Args otherwise as forecast_multivariate_<model> + per-slug weights.

• forecast_samples_ensemble

  Samples ensemble — concatenates sample paths from every supporting model along the sample axis. Per-model sample count is max(1, round(weight * num_samples)). median is recomputed over the pooled paths. Args: context, horizon, num_samples, context_length, unload: same as forecast_samples_<model>. weights: {slug: float >= 0}. None = uniform; weight 0 disables a model.

• forecast_univariate_ensemble

  Univariate ensemble — weighted mean across every model that supports univariate forecasting. Response carries the aggregated median and quantiles plus each contributing model's individual forecast and applied weight. Args: context, horizon, quantile_levels, context_length, unload: as in forecast_univariate_<model>. weights: {slug: float >= 0}. Missing slugs default to 1.0; weight 0 disables that model. None = uniform.

• unload_models

  Unload every resident foundation model and release model memory, Python garbage, Torch compiler caches, and CUDA cache resources. The operation refuses to run while any foundation forecast is active.

Signals

Stars

3

Forks

0

Open issues

0

Last commit

Oct 19, 2026 (10 days ago)

Last release

Oct 10, 2026 (19 days ago)

Last package publish

Oct 17, 2026 (12 days ago)

From GitHub (fetched Oct 24, 2026) and the package registry.

Details

Registry name

io.github.psyb0t/predictalot

Version

1.2.2

Auth

No auth

Packages

oci docker.io/psyb0t/predictalot:v1.2.2

License

WTFPL

Repository

github.com/psyb0t/docker-predictalot

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

Official MCP Registrygithub-graphql