AI Research Atlas

Claude Code added mods, FLUX 3 Image added bounding-box control, and Suno launched Speech

Daily brief · Thursday, 1 October 2026

Thursday. Written from the atlas logs (checked against sources 6 Oct 2026).

Claude Code gained "mods" written as small TypeScript functions, Black Forest Labs' FLUX 3 Image gained bounding-box layout control, and Suno released Speech in beta.

Claude Code "mods" let users, or Claude itself, change how the coding agent behaves with small TypeScript functions shipped in plugins, without forking it. Black Forest Labs' FLUX 3 Image added layout control by bounding boxes and edits that leave everything outside the box numerically unchanged, which matters for agents that edit images many times in a row. Suno Speech, in beta, generates voice and music together as one track.

Also in the news

  • Microsoft released streaming transcription and faster voice models for real-time voice agents.

Everything logged this day

  • FLUX 3 Image Black Forest Labs
    FLUX 3 Image adds JSON bounding-box layouts and edits that leave pixels outside the box numerically unchanged, native ~4K output, up to ten references.
  • Claude Code mods Anthropic
    Claude Code mods are small TypeScript functions that rewrite prompts, add UI or replace built-in features, shipped inside plugins.
  • Suno Speech (beta) Suno
    Suno Speech generates voice and music together as one track, e.g. a dramatic reading or meditation with original background music.
  • MAI-Transcribe-2-Streaming, MAI-Voice-2.1 and Voice-2.1-Flash Microsoft
    Streaming transcription ranks 1 on Artificial Analysis with first hypothesis in ~100 ms; Voice-2.1-Flash has 150 ms end-to-end latency.