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.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
open weights — A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.
Why it matters
Devstral 2 reaches 72.2% on SWE-benchThe 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.Full definition → at 123B parameters with 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 → under a modified MIT license. Devstral Small 2, at 24B parameters under Apache 2.0, runs on consumer hardware and offers a credible open weightsA model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.Full definition → option for self-hosted code agents.
