Skip to content

Commit 2f88b17

Browse files
committed
news 15
1 parent b4cb0b6 commit 2f88b17

File tree

1 file changed

+26
-0
lines changed

1 file changed

+26
-0
lines changed
Lines changed: 26 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,26 @@
1+
---
2+
title: Turing.jl Newsletter 15
3+
description: The fortnightly newsletter for the Turing.jl probabilistic programming language
4+
categories:
5+
- Newsletter
6+
author:
7+
- name: The TuringLang team
8+
url: /team/
9+
date: 2025-11-07
10+
---
11+
12+
**Faster model evaluation**
13+
14+
[This PR](https://github.com/TuringLang/DynamicPPL.jl/pull/1132) makes Turing faster by about 2-10x depending on the size of the model.
15+
(Not kidding, see the PR for examples!)
16+
Smaller models benefit the most, larger models less so because for those a larger proportion of your time is spent just calculating `logpdf`.
17+
18+
The main impact is probably in inference.
19+
The PR contains an example where NUTS sampling on eight-schools is cut from ~ 8 seconds to ~ 1.5 seconds.
20+
The general strategy should lead to speedups for other operations as well such as `returned` and `predict`, although those will be less marked as reading from / constructing a chain will be the main bottlenecks.
21+
22+
It's not released yet, but should hopefully be in the near future!
23+
24+
**Mooncake @ 1.12**
25+
26+
Just a note to say that there is now a newly released version of Mooncake that works on 1.12, and Turing should be perfectly compatible with it!

0 commit comments

Comments
 (0)