TuringLang/Turing.jl: v0.42.9

Kai Xu; Hong Ge; Cameron; David Müller-Widmann; Martin Trapp; Tor Erlend Fjelde; Mohamed Tarek; Penelope Yong; Markus Hauru; Will; Adam Scibior; Killian Q. Zhuo; Xianda Sun; Hessam Mehr; Emma Smith; Philipp Gabler; Jaime RZ; Harry Wilde; Aoife; Shravan Goswami; Kyurae Kim; azev77; Tom Röschinger; Pietro Monticone; FredericWantiez; Arthur Lui; hsm207; dfish; Saranjeet Kaur

doi:10.5281/zenodo.18837884

Published March 2, 2026 | Version v0.42.9

Software Open

TuringLang/Turing.jl: v0.42.9

1. University of Cambridge
2. @letta-ai
3. @PumasAI
4. Aalto University
5. @TuringLang
6. @alan-turing-institute
7. University of British Columbia
8. None
9. King's College London
10. UCL
11. @nestauk @dssg @turinglang @knowlab @BHFDSC
12. University of Pennsylvania
13. Fowler College of Business, SDSU
14. University of Trento

Turing v0.42.9

Diff since v0.42.8

Improve handling of model evaluator functions with Libtask.

This means that when running SMC or PG on a model with keyword arguments, you no longer need to use @might_produce (see patch notes of v0.42.5 for more details on this).

It also means that submodels with observations inside will now be reliably handled by the SMC/PG samplers, which was not the case before (the observations were only picked up if the submodel was inlined by the Julia compiler, which could lead to correctness issues).

Merged pull requests:

Run CI against new DPPL / VNT (#2756) (@penelopeysm)
CompatHelper: bump compat for Mooncake to 0.5 for package test, (keep existing compat) (#2758) (@github-actions[bot])
Tell users about the proposals that are being used in MH (#2774) (@penelopeysm)
Mark all methods with DynamicPPL.Model as produceable (#2780) (@penelopeysm)

Closed issues:

Time (#2763)
data[i, "γ (N/m)"] ~ Normal(γ_pred, σ) doesn't work upon v0.41 -> v0.42 (#2771)
error upon data[:, "y"] ~ MvNormal(...) in Turing v0.42.8 (#2773)

Files