Laguna S 2.1 gives teams building coding agents an open-weight mixture-of-expertsA model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.Full definition → alternative to closed frontier models, with a 1M-token context windowThe maximum amount of text a model can consider at once — its working memory for the current conversation or task.Full definition → for long-horizon repo work and toggleable thinking mode; the free tier on Vercel's AI Gateway means you can benchmark it against your own agent harnessThe 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.Full definition → before committing.
Terms in this piece · Glossary
mixture-of-experts — A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
SWE-bench — The standard benchmark for AI coding agents: real GitHub issues from real repositories, scored by whether the agent's patch passes the project's own tests.
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.
Key quotes
“Laguna S 2.1 is an open-weight Mixture-of-Experts model that supports a context window of up to 1M tokens and runs in thinking and no-thinking modes.”
“In thinking mode, Laguna S 2.1 reports 70.2% on Terminal-Bench 2.1, 78.5% on SWE-bench Multilingual, and 59.4% on SWE-Bench Pro.”
“AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests.”
