Wan 3.0 launches on Magnific with 20 reference support and native audio
Loo_Atreides · x · 2026-08-26
Wan 3.0 is now live on Magnific.
Key Updates:
- Multi-Reference Input: Supports 20 references per generation (10 images, 5 audio, 5 videos) instead of a single style reference.
- Length & Consistency: Generates 2-30 second videos in a single pass. No stitching or clip-byclip extension is required, preserving character consistency.
- Native Audio: Generates voice, ambience, and SFX simultaneously with the video, rather than overlaying them later.
- Adaptive Composition: Supports multiple ratios, with the model composing for the frame instead of cropping.
- Complex Control: Allows inputting up to 10 image references (e.g., character, product, location, lighting) to control a single shot.
A 40% discount is available until September 30.
More from Multimodal
- IBM releases Granite Speech 5.0 Turbo CTC for fast transcription — coder543 · 2026-08-26
- Google Nano model generates realistic Saudi man without reference images — aziz4ai · 2026-08-26
- MiniMax H3 draws a wine glass filled to the brim, where most models fail — nazihater3000 · 2026-08-26
- MiniMax H3 Lip Sync Flaw: Mouth Moves During Instrumental Passages, Measured and Workarounds Provided — Prestigious_Cat85 · 2026-08-26
- Reflecting on LLaVA: Teaching LLMs to see via visual encoder projection — alec_helbling · 2026-08-26
- Seedance 2.5 integrates with Claude Code, supports CLI — aliscodes · 2026-08-26