Generate images, video, and audio

Choose a generation mode

Open image, video, or audio generation in the AI tool and choose a provider and model supporting the task. Configure default generation providers in AI Settings or choose a model in the form. A chat model does not necessarily generate media.

Configure the request

For images, enter a prompt and add reference images or dimensions when supported. For video, use the available reference-media, keyframe, scene, or duration controls. For audio, provide text or the inputs required by the selected mode. Parameters vary by model.

Submit and save

Start with a small task and watch its progress, completion, or failure state. Tasks may complete asynchronously. Repeated submissions can create separate work and charges. Open completed results, save them locally, and check dimensions, duration, and audio. Remote delivery links can expire; keep important output as files.

Errors and usage

Check the provider, balance, parameters, and error before retrying. Use reference voices and assets only with appropriate authorization. See Account and privacy.