Modelos
Modelos de gerador de imagens IA e edição de imagens disponíveis no Create. Abra a página de um modelo para gerar com ele travado, ou use o espaço livre em /create
GPT Image 2
OpenAI’s newest image model for complex prompts, in-image typography, and photoreal product work — text-to-image or edit with up to 16 references.
Grok Imagine Image
xAI’s Grok Imagine model for fast text-to-image generation and prompt-based editing.
Nano Banana 2
Google’s fast Gemini image model for high-volume generation, multi-image editing, and sharp on-image text.
Nano Banana Pro
Google DeepMind’s flagship Gemini image model — studio-grade control, sharp text, and multi-image blending up to 4K.
Seedream 5 Pro
ByteDance's flagship image generation and editing model for sharp 1K and 2K outputs.
Seedream 5 Lite
Seedream 5.0 lite with built-in reasoning, example-based editing, and deep domain knowledge.
Z-Image Turbo
Tongyi-MAI’s 6B Z-Image Turbo for sub-second text-to-image generation.
Z-Image
Tongyi-MAI’s Z-Image foundation model with CFG and step control for text-to-image.
P-Image
Pruna’s sub-second P-Image model for production text-to-image generation.
P-Image Edit
Pruna’s multi-image editing model for fast prompt-based image changes.
FLUX.2 Klein 4B
Very fast FLUX.2 image generation and editing. Distilled for sub-second production use.
FLUX.2 Klein 4B Base
Un-distilled FLUX.2 Klein for flexible generation, editing, and post-training style control.
FLUX.2 Pro
High-quality FLUX.2 generation and editing with up to eight reference images.
FLUX.2 Max
Highest-fidelity FLUX.2 image model for generation and multi-reference editing.
FLUX.2 Dev
Quality FLUX.2 generation and editing with reference images and fast mode.
FLUX.2 Klein 9B
4-step distilled FLUX.2 Klein 9B for flexible, controllable generation and editing.
FLUX.2 Klein 9B Base
Un-distilled FLUX.2 Klein for flexible generation, editing, and post-training style control.
FLUX.2 Flex
Max-quality FLUX.2 generation and editing with up to ten references, steps, and guidance control.
Seedream 4.5
Upgraded ByteDance image model with stronger spatial understanding and world knowledge.
GPT Image 1.5
OpenAI image generation and editing with strong prompt following, readable text, and transparent background support.
Nano Banana
Google’s original Nano Banana model for fast generation, conversational edits, and multi-image fusion.
Seedream 4
Unified text-to-image generation and precise single-sentence editing at up to 4K resolution.
FLUX.1 Schnell
The fastest classic FLUX.1 model for quick text-to-image generation.
FLUX 1.1 Pro
Faster, higher-quality FLUX Pro with strong prompt adherence and output diversity.
FLUX.1 Kontext Pro
State-of-the-art text-based image editing with excellent prompt following and consistency.
FLUX.1 Kontext Dev
Open-weight FLUX.1 Kontext for text-based image editing with guidance and step control.
FLUX.1 Fill Pro
Professional inpainting and outpainting. Requires an image; optional mask and outpaint presets.
FLUX.1 Dev
12B rectified-flow FLUX.1 model for text-to-image and image-to-image generation.
FLUX.1 Fill Dev
Open-weight inpainting and extending. Requires an image; optional mask slot for masked edits.
FLUX 1.1 Pro Ultra
FLUX 1.1 Pro in ultra and raw modes. Up to 4 megapixel outputs with optional realism.
FLUX.1 Pro
Original FLUX.1 Pro with top-tier prompt following, detail, and output diversity.
FLUX.1 Kontext Max
Premium FLUX.1 Kontext editing with maximum performance and improved typography.
Seedream 3
A text-to-image model with native high-resolution (2K) image generation.
FLUX.1 Depth Dev
Open-weight depth-aware generation that preserves spatial relationships from a reference image.
FLUX.1 Canny Pro
Professional edge-guided image generation. Control structure and composition with a reference image.
FLUX.1 Depth Pro
Professional depth-aware image generation that preserves spatial relationships from a reference image.
FLUX.1 Canny Dev
Open-weight edge-guided generation using Canny structure from a reference image.
FLUX Krea Dev
Opinionated FLUX model with Krea that excels at photorealism and avoids the oversaturated AI look.