# 🚀 Release pymc-extras v0.16.0

**URL:** <https://discourse.pymc.io/t/release-pymc-extras-v0-16-0/17989>\
**Category:** News\
**Created:** [November 1, 2026, 11:18pm UTC](https://discourse.pymc.io/t/release-pymc-extras-v0-16-0/17989 "2026-11-01T23:18:24Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![pymc-bot](https://yyz2.discourse-cdn.com/flex036/user_avatar/discourse.pymc.io/pymc-bot/32/9384_2.png) [@pymc-bot](https://discourse.pymc.io/u/pymc-bot)\
**Post date:** [November 1, 2026, 11:18pm UTC](https://discourse.pymc.io/t/release-pymc-extras-v0-16-0/17989/1 "2026-11-01T23:18:24Z")

</div>

A new release of **pymc-extras** is now available!

## 📦 Release Information

- **Version:** `v0.16.0`
- **Repository:** [pymc-devs/pymc-extras](https://github.com/pymc-devs/pymc-extras)
- **Release Page:** [Release v0.16.0 · pymc-devs/pymc-extras · GitHub](https://github.com/pymc-devs/pymc-extras/releases/tag/v0.16.0)
- Note: It may take some time for the release to appear on PyPI and conda-forge.

## 📋 Release Notes

## What’s Changed

### New Features 🎉

- Fit BayesianETS through its innovations recursion instead of the Kalman filter by [@jessegrabowski](https://github.com/jessegrabowski) in [#777](https://github.com/pymc-devs/pymc-extras/pull/777)
- Implement conditional recovery for Laplace-marginalized variables by [@fonnesbeck](https://github.com/fonnesbeck) in [#790](https://github.com/pymc-devs/pymc-extras/pull/790)
- MLX MCLMC sampler by [@jessegrabowski](https://github.com/jessegrabowski) in [#743](https://github.com/pymc-devs/pymc-extras/pull/743)

### Bugfixes 🪲

- Fix GeneralizedPoisson scalar draw by [@ricardoV94](https://github.com/ricardoV94) in [#785](https://github.com/pymc-devs/pymc-extras/pull/785)
- Fix BetaNegativeBinomial generator parameterization by [@fonnesbeck](https://github.com/fonnesbeck) in [#787](https://github.com/pymc-devs/pymc-extras/pull/787)
- Fix GEV log-CDF at the finite upper endpoint by [@fonnesbeck](https://github.com/fonnesbeck) in [#788](https://github.com/pymc-devs/pymc-extras/pull/788)
- Preserve Prior core\_dims during serialization by [@fonnesbeck](https://github.com/fonnesbeck) in [#789](https://github.com/pymc-devs/pymc-extras/pull/789)
- Build the full-rank ADVI guide’s Cholesky factor with constant indices by [@jessegrabowski](https://github.com/jessegrabowski) in [#795](https://github.com/pymc-devs/pymc-extras/pull/795)

### Maintenance 🔧

- Fix forecast with exogenous data on custom statespace models by [@jessegrabowski](https://github.com/jessegrabowski) in [#794](https://github.com/pymc-devs/pymc-extras/pull/794)
- Run pathfinder on float32 models by [@jessegrabowski](https://github.com/jessegrabowski) in [#813](https://github.com/pymc-devs/pymc-extras/pull/813)
- Bump pymc to 6.4 and pytensor to 3.3.6 by [@jessegrabowski](https://github.com/jessegrabowski) in [#815](https://github.com/pymc-devs/pymc-extras/pull/815)

## New Contributors

- [@G26karthik](https://github.com/G26karthik) made their first contribution in [#802](https://github.com/pymc-devs/pymc-extras/pull/802)

**Full Changelog** : [Comparing v0.15.1...v0.16.0 · pymc-devs/pymc-extras · GitHub](https://github.com/pymc-devs/pymc-extras/compare/v0.15.1...v0.16.0)

* * *

_This post was automatically generated from the GitHub release._