Cognition
ModelsModels from this lab.2
RankedModels with an index.0
Top modelModel with the highest index.N/A
LatestRelease of the newest model.10 Sept 2026SWE-2
Best by category
Out of 100AgenticMulti-step tasks with tools.SWE-277.2
CodingCode writing and repair.SWE-277.7
ReasoningLogic problems and puzzles.—N/A
MultimodalTasks with images and text.—N/A
KnowledgeFacts and expert knowledge.—N/A
MultilingualTasks in many languages.—N/A
InstructionTasks with strict rules in the prompt.—N/A
MathMath problems.—N/A