In GSoC 2026First time in 2026

MLLAM

MLLAM is a collaborative community dedicated to advancing machine learning applications in weather forecasting, specifically for Limited Area Modeling (LAM). The organization hosts several open-source key repositories, including neural-lam, which focuses on neural weather prediction using graph-based models. Members are from various research institutions and national weather services committed to improve weather forecasting at regional, high-resolution scale.

Ideas list Website

Contributors in 2026

First cycle

Projects since 2016

2 a cycle on average

Cycles

1 of 11

First in 2026

Rank in 2026

#161

of 183, 0.2% of slots

Mentors

–

Not published

Contributors per cycle

Accepted projects each year, against the median organization that year

View as table

MLLAM contributors per cycle
	MLLAM	Median organization
2016	0	4
2017	0	4
2018	0	3
2019	0	4
2020	0	4
2021	0	4
2022	0	3
2023	0	4
2024	0	4
2025	0	5
2026	2	4

Stack and topics

Each one has its own page of organizations and trends

Machine learning AI Deep learning Earth science Weather forecasting

Quick answers

Is MLLAM in GSoC 2026?: Yes, with 2 contributors, in its first cycle.
How long has it taken part?: 1 of 11 cycles since 2016, first in 2026.
How many contributors does it usually take?: 2 a cycle on average, 2 in total. Its largest cycle was 2026, with 2.
What does it build with?: Python, NumPy, PyTorch, Xarray, Zarr.

Accepted projects

2 projects in 2026 · All 2 projects since 2016

Flexible Graph Construction for Neural Weather Prediction
Prajwal_07
Neural weather prediction models like neural-lam rely on graph neural networks, but the current graph construction pipeline is locked to rectangular...
Project
Generalizing to Probabilistic Forecasting Models
sir-sloth-the-lazy
The existing Neural-lam probabilistic forecasting model, GraphEFM, currently exists in an isolated branch (prob_model_lam) with a monolithic ARModel...
Project