Poolside expanded its Laguna XS.2 and M.1 coding models to a 256K context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition →, the amount of text a model can work with in one request. Poolside's update says M.1 was already being served at that size on its own API and on OpenRouter, while XS.2 would follow later the same day, with the updated configuration already on Hugging Face. The update's benchmarkA standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.Full definition → table is dated May 26, 2026.
These are the models from the original launch. Poolside's launch post, first published April 28, 2026, now carries a note that both models are served at 256K. The change concerns how much material each request can hold, not a new model.
Terms in this piece · Glossary
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
benchmark — A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.
agent harness — The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.
sandbox — An isolated environment where AI-generated code or agent actions run without being able to touch anything real.
