Live drop

dots3-note Preview

Dots StudioDropped Aug 14, 20263 builds3 makers1 prompt

Multimodal MoE for long-horizon agent workflows with native text/image/video/audio support.

The most interesting things people are building with AI, all in one place — followed and tracked as they evolve from first demo to real product.

1014 indexedPress / to search--:--:-- UTC

Dropday is new — we're looking for genuine feedback.

Home Leaderboard Top creators What is this?Privacy Terms

DROPDAY.ai

dots3-note Preview — Model Drop — Dropday.ai

dropday.ai

Home

Browse

Model Drops Leaderboard

← All model drops

Live drop

dots3-note Preview

Dots StudioDropped Aug 14, 20263 builds3 makers1 prompt

Multimodal MoE for long-horizon agent workflows with native text/image/video/audio support.

Browse in feed Submit a build

The most interesting things people are building with AI, all in one place — followed and tracked as they evolve from first demo to real product.

1014 indexedPress / to search--:--:-- UTC

Dropday is new — we're looking for genuine feedback.

Home Leaderboard Top creators What is this?Privacy Terms

DROPDAY.ai

About this release

What is dots3-note Preview?

dots3-note Preview is the first open-weight model in the dots3 family from Dots Studio (Xiaohongshu). It is a 280B-parameter multimodal MoE (16B active) with up to 512K context, supporting text, image, video, and audio inputs for tasks including agentic tool use, coding, reasoning, and long-context document/multimedia understanding. Released under Apache 2.0 with FP8 quantized variants for efficient local deployment.

Read the official Dots Studio launch page

Worth opening first