September models: Wan 3.0, GPT Image 2.5, Kling O3 and ElevenLabs v4
Video up to 30 seconds with sound, image generation with up to 16 references, Kling O3 with video transformation and the new ElevenLabs v4 voice-over.
September brought four model families to Clipia, for video, images and voice-over.
Wan 3.0, Wan 3.0 I2V and Wan 3.0 R2V
The new generation of the Alibaba video model: up to 30 seconds of continuous action in one generation, with sound generated together with the picture. I2V takes a first frame and optionally a last one, while R2V builds a clip from your own material: up to 10 images, 5 video clips and 5 audio tracks.
GPT Image 2.5 Flare and Sunburst
Image generation and editing with up to 16 references, 13 aspect ratios and 1K, 2K or 4K output.
Kling O3
Four modes of the Kuaishou video model: text to video, image to video, transformation of a source clip and reference to video. Up to 15 seconds, character references and native audio.
ElevenLabs v4 and v4 Turbo
New voice-over from ElevenLabs in the audio studio: it reads a script with context and delivery cues, handles a two-voice dialogue and speaks Russian and other languages. Turbo is the fast option.
How to try them
Open the video studio, the image studio or the audio studio and pick the model from the list.
Try this model right now — no setup required.
Try Kling O3
