Guides, model notes and product updates

A corrected walkthrough of the TokenLab image edit endpoint: the right path, the two accepted image input shapes, multi-image support on gpt-image-2, and how async polling works when an edit is slow.

TokenLab now supports Kling 3.0 element references in the video generation API, letting developers anchor multiple products, props, or characters with @name tags. We cover the request pattern, the no-audio constraint, and the implementation checklist.