AI image models

Kyeo AI's Z-Image workflow is a streamlined text-to-image option. The public source service charges a raw price of 0.8 credits per image, and the local integer-credit wallet rounds it up to 1 credit; prompts can contain up to 1,000 characters, and 5 aspect ratios are available. The workflow does not accept image input or provide editing.

Kyeo AI offers text-to-image generation and reference-guided editing with Nano Banana 2. You can upload up to 14 reference images, choose from 15 aspect ratios, and output at 1K, 2K, or 4K. Those tiers currently cost 5, 8, and 12 credits, prompts allow 20,000 characters, and the current generation form does not offer web search.

Nano Banana Pro supports text-to-image generation and reference-image editing on Kyeo AI, with 1K, 2K, or 4K output. Each 1K or 2K request costs 8 credits, while 4K costs 14. Resolution is an output-size tier, not a guarantee of quality, accuracy, or detail.

Nano Banana, also known as Gemini 2.5 Flash Image, provides text-to-image generation and natural-language editing on Kyeo AI. Requests without images use `Nano Banana`; requests with images use `Nano Banana`. Both paths cost 3 credits and accept prompts up to 5,000 characters.

GPT Image 2 / Capabilities — GPT Image 2 supports text-to-image generation and reference-image editing on Kyeo AI. Kyeo charges by output resolution: 1K costs 3 credits, 2K costs 5, and 4K costs 8.

GPT Image 1.5 on Kyeo supports text-to-image and image-to-image with three aspect ratios and medium/high quality. Image editing accepts up to 16 JPG, PNG, or WebP references at 10 MB each. Medium costs 4 credits and high costs 22 credits. GPT Image 1.5 is OpenAI's deprecated previous image model. Kyeo retains generation and editing access, but that availability does not make it OpenAI's current default choice.

4o Image / Capabilities — This page offers only three aspect ratios—`1:1 / 3:2 / 2:3`—at a fixed 3 credits per request and does not accept reference images.

Seedream 4.5 / Capabilities — ByteDance's official Seedream 4.5 material highlights multi-image editing, preservation of reference details, layout, and text rendering. The Kyeo workflow has its own concrete limits: prompts can contain up to 3,000 characters, 8 aspect ratios are available, and editing accepts up to 14 JPG, PNG, or WebP images at 10 MB each. Kyeo AI currently charges a fixed 7 credits per request, with the same price for basic and high. Budget for retries, revisions, and the share of outputs that pass review.

Seedream 5.0 Lite / Capabilities — ByteDance's lighter Seedream image model with shared pricing across text-to-image and image editing. It supports 8 aspect ratios, up to 14 reference images, and currently costs 6 credits per run. Basic 2K / High 3K / Ultra 4K · PNG / JPEG · NSFW checker.

Grok Imagine on Kyeo AI covers both text-to-image and one-image editing. A request without an upload uses text-to-image mode; attaching one image switches to one-image editing mode. Text-to-image returns 2 images for 4 credits in standard mode or 4 images for 5 credits in quality mode, while editing costs 4 credits per image and does not show that switch.

Flux 2 Pro / Capabilities — Flux 2 Pro — Text-to-image and editing use the same resolution pricing: 5 credits for 1K or 7 credits for 2K. 3–5,000 characters; Up to 8 images.

Flux 2 Flex supports prompt-only generation and editing with up to eight images on Kyeo. A request without images uses Flux 2 Flex; an upload switches it to Flux 2 Flex. Seven fixed aspect ratios and 1K or 2K output are available for 14 or 24 credits per request.

Flux Kontext Max provides prompt-only generation and one-image editing on Kyeo for a fixed 8 credits. It exposes the same controls as Pro: generation offers six aspect ratios, while editing accepts one image and hides the aspect-ratio control.

Flux Kontext Pro provides prompt-only generation and one-image editing on Kyeo for a fixed 4 credits. Generation offers six aspect ratios. Uploading one image switches to editing, where Kyeo hides the ratio control and sends no ratio by default. The page also exposes JPEG or PNG, automatic translation, prompt enhancement, safety tolerance, and a watermark identifier.

Qwen2 Image — Qwen2 Image / Text creation and one-image editing — Start from a text prompt or upload exactly one image to edit. Multiple reference images are not supported. Ratio, format, and seed controls — Text creation supports 1:1, 3:4, 4:3, 9:16, and 16:9. Editing adds 2:3, 3:2, and 21:9; the default remains 16:9. Prompts are capped at 800 characters; editing accepts one JPG, PNG, or WebP image up to 10 MB; each result costs a fixed 6 credits.

Qwen Image supports prompt-only generation and one-image editing on Kyeo, with six generation sizes, PNG or JPEG output, prompts up to 5,000 characters, and 4 credits per output megapixel.

Recraft Crisp Upscale is Kyeo's one-image clarity and resolution enhancement action. It accepts one JPG, PNG, or WebP file up to 10 MB, exposes no scale, output-format, or other control, and costs a fixed 1 credit. Compare downloaded files to determine the actual dimensions, sharpening, and detail changes.

Recraft Remove Background / Capabilities — Kyeo accepts one JPG, PNG, or WebP image up to 5 MB and charges a fixed 1 credit. Download the result to review transparency and subject edges. 1 JPG, PNG, or WebP image up to 5 MB and 16 MP; each dimension must be 256–4096 px.

Wan 2.7 Image Pro / Capabilities — Wan 2.7 Image Pro supports text-to-image generation and image editing with up to nine reference images. Fixed 12 credits per image; each request returns one image.

Wan 2.7 Image / Capabilities — Wan 2.7 Image supports text-to-image generation and image editing with up to nine reference images. Fixed 5 credits per image; each request returns one image.

Text-only image generation. 12 credits per request. 5,000 characters. Aspect ratio: 1:1, 16:9, 9:16, 3:4, 4:3, Auto.

Text-only image generation. 8 credits per request. 5,000 characters. Aspect ratio: 1:1, 16:9, 9:16, 3:4, 4:3, Auto.

Text-only image generation. No image input. Aspect ratio: 1:1, 16:9, 9:16, 3:4, 4:3, Auto. 4 credits per request.

Topaz Image Upscaler / Capabilities — Topaz Image Upscaler accepts one image and one scale factor on Kyeo AI. It does not read a prompt or change the subject or composition. The values 1, 2, and 4 are width-and-height multipliers, not fixed 1K, 2K, or 4K outputs; calculate the output by multiplying the source width and height by the selected factor.

Ideogram V3 is Kyeo AI's design-first AI image generator for posters, ad layouts, cover art, and social graphics where text rendering and typography matter.

Ideogram V3 Remix — Upload one JPEG, PNG, or WebP image, describe what should stay or change, then choose speed, style, MagicPrompt, size, strength, and 1–4 outputs. It suits experiments with style, setting, material, or content based on an existing poster, illustration, product concept, or social asset.

Character is Ideogram V3's character-reference generation path. The reference anchors who the subject is while the prompt defines the new image; it is neither fidelity-preserving editing nor multi-person identity management. 12 / 18 / 24 credits per image; 1–4 outputs

Seedream 5.0 Pro combines text-to-image generation with editing from up to 10 reference images in one entry point. basic maps to 1K and high to 2K. Editing preauthorization rises with the number of inputs, making it useful for multi-image workflows that need a known cost before submission.

Seedream 5.0 Pro Layer Decomposition is designed for one existing image. It produces a base image and exports detected elements as transparent PNG layers. Kyeo validates the true file type, total pixel count, and aspect ratio before the asset enters the task pipeline.

Nano Banana 2 Lite is a streamlined image entry point: generate from text alone or upload up to 10 reference images, choose only the aspect ratio, and use a fixed 3 credits per request. There is no resolution button, which prevents a pricing-table label from being presented as an interface capability.

Qwen Image 3.0 supports both text-only generation and image-to-image work with up to 3 references. It offers 1K/2K, 8 canvas sizes, PNG/JPEG output, prompt expansion, negative prompts, and seed control. Kyeo preauthorizes 5–7 credits per request.

Qwen Image 3.0 Pro offers text-to-image generation and image to image with up to 3 references. It includes 1K/2K, 8 canvas sizes, PNG/JPEG output, prompt expansion, negative prompts, and seed control. Preauthorization ranges from 7–14 credits based on resolution and input count.

Grok Imagine Image 2.0 combines prompt-only generation and standard image editing with up to 5 references in one on-site entry. Text to image offers 5 fixed ratios, while editing also supports auto; both available modes currently reserve 4 credits per image.