Models
uooki has no model picker. Name a model in the conversation and the agent routes to it; leave it out and it chooses based on what you asked for. These pages are about when naming one is worth it.
Image15
Flux 2 Pro
Black Forest Labs (FLUX.2 Pro)
BFL's house aesthetic — visibly a different lineage from the Google and OpenAI models.
per image→Flux Kontext Max
Black Forest Labs (FLUX.1 Kontext Max)
The high end of the edit tier, for edits Kontext Pro cannot hold together.
per image→Flux Kontext Pro
Black Forest Labs (FLUX.1 Kontext Pro)
The standard tier for image **editing** — bring an image, change one thing, leave the rest alone.
per image→GPT Image 2
OpenAI
uooki's default image model — the one you get when you don't name one.
per image→Grok Image
xAI (Grok Imagine)
xAI's visual lineage, at a low price per image.
per image→Grok Image Edit
xAI (Grok Imagine Edit)
The cheapest editing route in the catalog.
per image→Imagen 4.0
Google (Imagen 4.0)
The photoreal-leaning branch of the catalog.
per image→Nano Banana
Google (Gemini 2.5 Flash Image)
The strictest reference adherence in the catalog — use it for many images of the same thing.
per image→Nano Banana 2
Google (Gemini 3.1 Flash Image)
Google's current default image model — Pro-tier quality on the Flash-tier clock.
per image→Nano Banana 2 Lite
Google (Gemini 3.1 Flash Lite Image)
The same price as the original Nano Banana, one generation newer.
per image→Nano Banana Pro
Google (Gemini 3 Pro Image)
The best instruction-following in the catalog — the more specific your brief, the further ahead it pulls.
per image→Qwen Image 2.0
Alibaba (Qwen Image 2.0)
The strongest Chinese, Japanese and Korean text rendering here — nothing else in the catalog replaces it.
per image→Seedream 5.0 Lite
ByteDance (Seedream 5.0 Lite)
The Seedream family's budget tier, and the cheapest model that still reads as finished work.
per image→Seedream 5.0 Pro
ByteDance (Seedream 5.0 Pro)
4K output and Chinese-language scene understanding in the same model.
per image→Z Image Turbo
Tongyi-MAI
The cheapest model in the catalog, built for volume.
per image→
Video12
Grok Video
xAI (Grok Imagine Video)
xAI's video lineage, with reference-image support.
per second→Hailuo 2.3
MiniMax (Hailuo 2.3)
The cheapest model here that still produces something you would show someone.
per second→HappyHorse 1.1
Alibaba (HappyHorse 1.1)
Ranked first in the image-to-video arena, with lip-sync built in.
per second→Kling O3
Kling (O3)
A strict superset of Kling V3 — same price when silent, more capability.
per second→Kling V3
Kling (V3)
Ranked first in the text-to-video arena.
per second→PixVerse V6
PixVerse (V6)
The cheapest video tier in the catalog.
per second→Seedance 2.0
ByteDance (Seedance 2.0)
Takes more reference material than anything else here — nine images plus a reference video.
per second→Veo 3.1
Google (Veo 3.1)
The flagship of the catalog, and native audio is what separates it from everything else here.
per clip→Veo 3.1 Fast
Google (Veo 3.1 Fast)
The middle of the Veo family — still the Veo look, at a fraction of the cost.
per clip→Veo 3.1 Lite
Google (Veo 3.1 Lite)
The cheapest way into the Veo family.
per clip→Vidu Q3
Vidu (Q3)
Multi-reference consistency, up to four images.
per second→Wan 2.6
Alibaba (Wan 2.6)
Alibaba's open-source lineage, with motion that reads differently from the Kling and Seedance families.
per second→