AI video and image models on CraftyPeak

Seedance 2.5

Multi-reference video with native sound, directed from one prompt.

From 13 credits a second ($0.13)

Seedance 2.0

Cinematic multi-shot video with sound, from drafts to 4K.

From 9 credits a second ($0.09)

Seedance 2.0 Fast

Seedance 2.0 at speed, for previews, ideas and volume.

From 5 credits a second ($0.05)

Seedance 2.0 Mini

The compact Seedance for drafts, loops and bulk video.

From 3 credits a second ($0.03)

Seedance 1.5 Pro

Dependable Seedance video, with sound on or off.

From 4 credits a second ($0.04)

Veo 3.1

Google's Veo 3.1: realistic video with matching sound.

From 26 credits a second ($0.26)

Veo 3.1 Fast

The Veo 3.1 look, faster and for fewer credits.

From 11 credits a second ($0.11)

Veo 3.1 Lite

The entry tier of Veo, for drafts and volume.

From 4 credits a second ($0.04)

Google Omni Flash

Create video from text, or edit your footage with a prompt.

From 14 credits a second ($0.14)

Wan 3.0

One video model for text, reference images, frames and video edits, with sound.

From 5.5 credits a second ($0.06)

Wan 3.0 Prime

Wan 3.0, tuned for speed.

From 7.48 credits a second ($0.07)

Wan 2.7

Text, reference images, start and end frames, and video edits in one model family.

From 11 credits a second ($0.11)

Wan 2.6

Multi-shot video from text or reference images, with synced sound.

From 2.75 credits a second ($0.03)

HappyHorse 1.1

More expressive motion and steadier characters, with sound.

From 7.7 credits a second ($0.08)

HappyHorse 1.0

The video model that debuted at the top of the arena.

From 15.4 credits a second ($0.15)

Nano Banana Pro

Studio-grade images with legible text, up to 4K.

From 18 credits an image ($0.18)

Nano Banana 2

Pro-style images at Flash speed.

From 6 credits an image ($0.06)

Nano Banana

The original Nano Banana for everyday edits.

From 6 credits an image ($0.06)

Seedream 5.0 Pro

Photoreal detail and design-ready layouts.

From 5 credits an image ($0.05)

Seedream 5.0 Lite

Large, detailed images at a small price.

From 4 credits an image ($0.04)

FLUX.2 [dev]

Open-weight FLUX.2 for detailed images and reference edits.

From 3 credits an image ($0.03)

FLUX.2 [klein] 9B

Compact FLUX.2 with more detail than 4B.

From 2 credits an image ($0.02)

FLUX.2 [klein] 4B

The quickest, smallest FLUX.2.

From 1 credit an image ($0.01)

FLUX.1 [schnell]

The classic quick FLUX for instant drafts.

From 1 credit an image ($0.01)

SDXL Lightning

Stable Diffusion XL in a few quick steps.

From 1 credit an image ($0.01)

Stable Diffusion 1.5

Classic Stable Diffusion restyling for your images.

From 1 credit an image ($0.01)

Wan 2.7 Image Pro

The higher-quality Wan 2.7 Image tier, for generation and editing.

From 9 credits an image ($0.09)

Wan 2.7 Image

Generate and edit with up to nine input images, with text in twelve languages.

From 4 credits an image ($0.04)

Wan 2.6 Image

Edit with up to four images, or make new images with Wan 2.6 T2I.

From 4 credits an image ($0.04)

Qwen-Image 3.0

Infographics, UI layouts and ten-pixel text in one pass.

From 4 credits an image ($0.04)

Qwen-Image 2.1

Generation and editing with ten references and native transparency.

From 2 credits an image ($0.02)

Qwen-Image 2.0

Detailed text to image, with up to six results per request.

From 4 credits an image ($0.04)

Z-Image Turbo

Quick, low-cost text to image with Chinese and English lettering.

From 2 credits an image ($0.02)

Kling 4.0

Thirty-second, multi-shot video with stereo sound and fifteen references.

Kling 4.0 Flash

The quicker, lighter tier of Kling 4.0.

Kling 3.0

Multi-shot storyboards with characters that hold steady and speak.

Kling 3.0 Omni

One model for references, video editing and voices.

Kling 3.0 Turbo

Kling 3.0 speed with synced sound.

Kling O1

Generate and edit video in plain language, no masks.

Kling 2.6

Picture, voice and sound effects in one pass.

Kling 2.5 Turbo

Dynamic motion and camera work at a lower cost.

Wan 2.5

The first Wan with native, lip-synced sound.

Wan 2.2

Open-weight cinematic video from a mixture of experts.

Wan 2.1

The open Wan that writes Chinese and English text in video.

Wan 2.2 Animate

Copy a performance, or swap a character into a video.

Wan 2.2 Speech-to-Video

Animate a portrait to your own voice track.

Wan 2.1 VACE

One model for creating and editing video.

MiniMax H3

Twelve mixed references, motion transfer and stereo sound.

Hailuo 2.3

Complex body motion, micro-expressions and stylised looks.

Hailuo 2.3 Fast

Hailuo 2.3 image to video for batches.

Hailuo 02

Extreme physics and sharp instruction following.

Vidu Q4

Lifelike acting, from fifteen references.

Vidu Q3

Sixteen seconds of video and sound, multi-shot.

Vidu Q2

Acting-led video with correct Chinese text.

Vidu Q3 Turbo

The quicker Vidu Q3 tier.

Vidu Q2 Pro

Seven references and video editing.

PixVerse V6

Accurate camera moves and emotions that carry across cuts.

PixVerse V5.5

Dialogue with lip sync, music and multi-part video.

PixVerse V5.6

The final V5 update: cleaner visuals and vocals.

Luma Ray3.2

Sixteen keyframes and HDR for professional pipelines.

Luma Ray3.14

Native full HD, faster and lighter than Ray3.

Luma Ray3

Luma's reasoning video model with HDR.

LTX-2.5 Pro

Open-weight video with sound, multi-shot and high resolution.

LTX-2.3 Pro

Retake, extend and native portrait video.

LTX-2.5 Fast

Up to twenty seconds, quickly.

LTX-2.3 Fast

Sharper, faster LTX-2.3 for portrait and landscape.

Nano Banana 2.1

Fourteen references, search grounding and panoramic ratios.

Nano Banana 2 Lite

Images in about two seconds.

Kling Image 3.0

Reference-faithful images and precise edits from Kling.

Kling Image 3.0 Omni

Consistent image series, relighting and virtual try-on.

Kling Image O1

Edit images in plain language, or doodle the change.

Qwen-Image

Open-weight images with precise Chinese and English text.

Qwen-Image-Edit

Open-weight image editing from a prompt.

Qwen-Image-Layered

Split any image into editable layers.

Qwen-Image-2512

More realistic people, textures and text.

Qwen-Image-Edit-2509

Multi-image editing with built-in ControlNet.

Qwen-Image-Edit-2511

Consistent characters, group photos and new viewpoints.

GPT Image 2.5 Sunburst

OpenAI's image model for precise generation and editing.

GPT Image 2.5 Flare

Fast, high-quality everyday images from OpenAI.

GPT Image 2

OpenAI's first image model that reasons.

GPT Image 1.5

Sharper instruction following and precise edits.

FLUX 3 Image

Control over every pixel, with layout boxes and ten references.

FLUX 3 Image Klein

Open-weight FLUX 3 that runs on a consumer GPU.

FLUX.2 [max]

The top FLUX.2 tier, with web grounding.

FLUX.2 [pro]

Production-grade FLUX.2 at an affordable price.

FLUX.2 [flex]

Tunable FLUX.2 for typography and fine detail.

FLUX.1 Kontext [max]

Instruction-based editing with strong typography.

FLUX.1 Kontext [pro]

Fast instruction-based editing in seconds.

FLUX1.1 [pro] Ultra

Four-megapixel images and a candid Raw mode.

FLUX1.1 [pro]

Reliable, fast FLUX for everyday images.

Stable Diffusion 3.5 Large

The flagship of Stable Diffusion 3.5.

Stable Diffusion 3.5 Large Turbo

SD 3.5 Large in four steps.

Stable Diffusion 3.5 Medium

Compact SD 3.5 for consumer hardware.

Seedream 4.5

Dense typography and consistent multi-image edits.

Seedream 4.0

One model to generate, edit and batch at high resolution.

Ideogram 4.5

Precise, mask-based edits from Ideogram.

Ideogram 4.0

Layout boxes, hex palettes and JSON prompts for text-led design.

Ideogram 3.0

Style references, character reference and layered output.

Recraft V4.1

Design-led images and precise vectors.

Recraft V4

Design-led raster images from Recraft.

Recraft V4 Vector

Native, editable SVG from a prompt.

Recraft V4 Pro

Larger, more detailed Recraft V4 images.

Recraft V4 Pro Vector

Detailed, editable SVG from Recraft V4 Pro.

MiniMax image-01

Portrait-consistent images in batches of nine.