Overview
Field: AI Authors: Julien Abadji, Marah Abdin, Connor Adams, et al. Published: 2026-05-28 arXiv: 2605.27605
Summary
This report presents Laguna M.1 and Laguna XS.2, two Mixture-of-Experts (MoE) foundation models built for long-horizon, agentic coding:
- Laguna M.1: 225.8B total parameters (23.4B activated per token)
- Laguna XS.2: 33.4B total parameters (3B activated per token)
- The paper describes the principles and design choices behind the Model Factory.
- It details the full end-to-end training pipeline: pre-training data and architecture, post-training stages, evaluation, and quantization.
- On agentic software engineering and terminal benchmarks — SWE-bench Verified, SWE-bench Multilingual, SWE-Bench Pro, and Terminal-Bench 2.0 — M.1 and XS.2 are competitive with state-of-the-art open-weight models at their respective size classes.
- Laguna XS.2 weights are open-sourced under the Apache 2.0 license on Hugging Face.
Both models were trained from scratch, end-to-end, inside the authors' internal system called the Model Factory: a tightly integrated stack of versioned data, training, evaluation, and inference components that turns model development into an industrial process.
Key Points
Abstract (excerpt)
> We present Laguna M.1 and Laguna XS.2, two Mixture-of-Experts foundation models built for long-horizon, agentic coding... Both models were trained from scratch end-to-end inside the same internal system that we refer to as our Model Factory: a tightly-integrated stack of versioned data, training, evaluation, and inference components that turn model development into an industrial process...
*Auto-collected on 2026-05-29.*