New image, video and audio models

A batch of new models across images, video, voice and music. Pick them from the model picker on any board.
Images
- Ideogram 4.5, made for clean text and typography in images.
- Qwen Image 3 and Recraft V4.1.
Video
- Wan 3.0 Prime, LTX 2.5 Pro, LTX 2.5 Fast, Kling 3.0 Turbo Pro and PixVerse V6.
- Gemini Omni Flash 1.1 renders up to 4K with sound, and takes up to 10 reference images and 3 reference videos.
- Grok Imagine Video 1.5 makes clips from 1 to 15 seconds.
- MiniMax H3 Max can now generate from references.
Voice and music
- ElevenLabs v4 is the new default voiceover model.
- Gemini 3.8 Flash TTS adds 30 more voices.
- Google Lyria 3.5 makes full songs, and ElevenLabs Music v2.5 joins the music models.
A few older models made way for their newer versions and no longer appear in the picker.



