Introducing Laguna S 2.1: A High-Performance Mixture-of-Experts Model for Long-Horizon Coding Tasks

Laguna S 2.1 is a 118B parameter Mixture-of-Experts model with 8B activated parameters per token and 1M token context window, achieving strong performance on long-horizon coding benchmarks. It outperforms larger models in specific metrics while providing full evaluation trajectories for transparency.

Comments

Read original article