TypeSafe AI released Jev 1.14, adding experimental vision and Box positioning capabilities
TypeSafe AI released Jev 1.14, supporting image input, object bounding-box positioning, and approximately 262000 Token context, and is expected to October 21 become the default stable version.