Home / Models / Dots3-Note Preview

Dots3-Note Preview

Maker: Xiaohongshu · reasoning

First open-weight model in the dots3 series: a 280B mixture of experts with 16B active parameters and a 512K context, accepting text, image, video and audio. Introduces TEMPO, a reinforcement learning method for long-horizon agent tasks.

выход 14.08.2026. Производитель — Dots Studio, лаборатория ИИ внутри Xiaohongshu (RedNote), ранее выпускала dots.llm и dots.ocr под именем rednote-hilab. Ни Xiaohongshu, ни Dots Studio в справочнике нет

Understands

text, images, audio, video

Produces

text, code

Contents verified with the vendor: 2026-08-18

Included in subscriptions

No catalogue entry marks this model as part of a plan yet. It will appear here once the plan contents are confirmed with the vendor.