Every image and video model, one place
Veo 3.1, Sora 2, Kling 3, FLUX.2, Nano Banana 2 and more in Axe AI Studio. What each one is good at, and what it costs, before you run it.
Video models
- Veo 3.1
Veo 3.1 is the highest quality video model in Axe AI Studio, and it makes sound with the picture. Veo 3.1 Fast is quicker and cheaper, and can start from a frame. 10.4 credits a second (about $0.52).
- Sora 2
Sora 2 is strong at storytelling shots: a moment with a beginning and an end. It starts from text or from a frame you already have, and runs up to 12 seconds. 2.6 credits a second (about $0.13).
- Kling 3
Kling 3 is the pick for realistic people and action, when the people in the shot have to look and move like real ones. 4.42 credits a second (about $0.22).
- Seedance 2.0
Seedance 2.0 brings cinematic camera moves. It can start from text, or from a frame and keep that frame's look, which is how Studio keeps a film consistent shot to shot. 3.12 credits a second (about $0.16).
- Wan 2.7
Wan 2.7 is good value with smooth motion. Use it to try ideas quickly, or to animate a frame you already like. 2.6 credits a second (about $0.13).
Image models
- FLUX.2
FLUX.2 Pro gives sharp detail and great lighting. FLUX.2 Klein is fast and free every day, so you can draft ideas before you spend a credit. 1.04 credits an image (about $0.052).
- Nano Banana 2
Nano Banana 2 is the best at matching reference photos. Add up to three, of a person, a product or a place, and it keeps them recognisable in the new image. 2.08 credits an image (about $0.10).
- GPT Image 2
GPT Image 2 follows complex prompts closely: many objects, exact positions, words on signs. GPT Image 1.5 is there too. 5.2 credits an image (about $0.26).
- Seedream 5 Pro
Seedream 5 Pro makes cinematic frames and keeps faces consistent from one image to the next. It is Studio's default pick for storyboard frames. 1.3 credits an image (about $0.065).
- Qwen Image 2
Qwen Image 2 is strong at text and layouts: menus, posters, labels and slides that read cleanly. 1.3 credits an image (about $0.065).