Media workflows

Start with an idea. Leave with something you can use

Image, video, audio, or 3D — each opens in Console with a prompt and model already set.

Featuredgpt-image-2

Text to image

Make an image from a prompt.

Recommended modelUse gpt-image-2 to generate images from text prompts.

  1. 01

    Choose

    Image, video, audio, music, 3D, or worlds.

  2. 02

    Create

    The prompt and model are already filled in.

  3. 03

    Keep

    Keep the work you like, along with its prompt and settings.

/dashboard/create?target=media&workflow=text-to-image&model=gpt-image-2&operation=image_generation

Ways to create

Ready to try

Every option arrives with a working model, prompt, and settings you can change.

Image & Video

Edit image

seedream-4.5

Edit an uploaded image with a prompt.

Text to video

seedance-2.0-fast

Turn a prompt into video.

Image to video

seedance-2.0-fast

Bring an uploaded image to life.

Voice & Audio

Text to speech

qwen3-tts-flash

Turn written text into speech.

Speech to text

fun-asr-flash-2026-06-15

Turn an audio recording into text.

Text to music

suno-music

Write original music from a prompt.

Worlds & 3D

Text to 3D

tripo-h3.1

Turn a prompt into a 3D model.

World generation

marble-1.1-plus

Generate an explorable 3D world or scene.

Sign in to create — every job shows its price first.

Earlier options

Existing integrations

Still here for integrations that already use them. Start new projects on a current option.

Image variation

gpt-image-1

For existing gpt-image-1 integrations. New projects should use Edit image, which runs seedream-4.5.

Audio translation

whisper-1

For existing whisper-1 integrations. New projects should use Speech to text and leave translation to a chat model.