Create AI Images with Z-Image, Krea2 Turbo, Anima, and WAN 2.1
Choose Z-Image, Krea2 Turbo, Anima, or WAN 2.1 to create AI images, anime artwork, fashion visuals, and image-to-video motion clips.
- Z-Image and Krea2 Turbo support realistic and editorial image generation.
- Anima supports anime-style illustration workflows.
- WAN 2.1 turns source images into short motion clips.
- Queue estimates help set expectations before generation starts.
Which model should you pick?
Use Z-Image Turbo when you want a fast realistic or editorial image and you are still iterating on the prompt. Use Krea2 Turbo when you want a more polished render or one of the optional LoRA character presets. Use Anima for anime-style illustration, and WAN 2.1 when you already have a finished image and want a short motion clip from it.
If you are not sure how to word a prompt, upload a reference photo to PromptLAB first and start from the structured prompt it returns. If a result is close but soft, run it through the AI Upscaler instead of regenerating it.
What a generation costs and how the queue works
Every job is reserved against a credit balance when it is queued: 2 credits per image for Z-Image Turbo, Krea2 Turbo and Anima, and 10 credits for a WAN 2.1 clip. Free credits refill every day (5 for guests, 20 when signed in; unused credits do not carry over), and the credits for a job that fails are returned automatically.
Jobs run on a shared queue, so the estimate shown before you submit is a guess based on recent run times, not a promise.
Where to learn from real results
The Articles section documents prompt tests we actually ran, including the failures: what a model ignored, where hands or text broke, and which wording helped. Read a guide for the model you chose before spending credits on long prompt experiments.