AFFOGATO MODELS

Every model, one workspace.

185 image, video and voice models from 36 makers, each with what it does and what a run costs in credits. Pick one in the studio, or call it from the API.

01IMAGE

Image models (102)

From Affogato, Alibaba, BiRefNet, Black Forest Labs, Bria AI, ByteDance, ClarityAI, fal, Finegrain, Google, HiDream, Ideogram, Krea, Kuaishou, OpenAI, Real-ESRGAN, Recraft, rembg, SoulX, Tencent, Topaz Labs, xAI. Try them in the AI Image Generator.

ModelMakerWhat it doesPrice

Affogato Realism

AffogatoMakes an image read as a real photo, picking the best realism engine for each prompt.27 cr

BiRefNet

BiRefNetHigh-quality background removal with crisp, accurate edges.4 cr

ByteDance Seedream v4.5

ByteDanceUnified generation + editing. Output up to 4K resolution.8 cr

ByteDance Seedream v4.5 Edit

ByteDanceEditing endpoint with multi-image inputs.8 cr

Clarity Upscaler

ClarityAIDetail-preserving creative upscaler popular for portraits and illustration.9 cr

Clothes Swap

AffogatoDresses the person in your photo in one or more garments you upload.15 cr

Face Swap

AffogatoSwaps up to three faces into a photo, with optional face restoration.20 cr

FLUX.1 [dev]

Black Forest LabsHigher quality than Schnell with 28 default inference steps. Supports character reference and ControlNet variants.5 cr

FLUX.3 Image

Black Forest LabsBlack Forest Labs' FLUX.3 Image — the flagship FLUX text-to-image model: photoreal detail, reliable typography, native 1K output.10 cr

FLUX.3 Image Edit

Black Forest LabsFine-tune image details with a written instruction while preserving the rest of the scene.5 cr

GPT Image 2.5 Flare

OpenAIOpenAI's GPT Image 2.5 Flare — GPT Image 2.5 quality at up to half the latency; native 4K.11 cr

GPT Image 2.5 Flare Edit

OpenAIOpenAI's GPT Image 2.5 Flare editing model: fast reference-guided edits.11 cr

Krea 3 Edit

KreaKrea 3 editing model: reference-guided edits and restyles with up to 8 input images.10 cr

Nano Banana 2.1

GoogleGoogle's Nano Banana 2.1 — the Flash-tier Gemini image model: sharper text, stronger character consistency, 1K to 4K output.8 cr

Nano Banana 2.1 Edit

GoogleGoogle's Nano Banana 2.1 editing model: multi-image references, in-place edits, 1K to 4K output.9 cr

Nano Banana Pro

GoogleState-of-the-art high-fidelity generation with strong typography and realism.30 cr

Nano Banana Pro Edit

GoogleEditing — high-fidelity edits with strong typography and realism.30 cr

Recraft Crisp Upscaler

RecraftCrisp upscaler for clean line art and graphic design.4 cr

Seedream 5.0 Flash

ByteDanceByteDance's Seedream 5.0 Flash — the speed tier of Seedream 5: Pro-class composition at a fraction of the cost, 1K and 2K output.6 cr

Seedream 5.0 Pro

ByteDanceByteDance Seedream 5.0 Pro — flagship text-to-image with deep-thinking prompt understanding, native text rendering in 14 languages, and precise control over dense layouts and structured design.27 cr

Seedream 5.0 Pro Edit

ByteDanceByteDance Seedream 5.0 Pro image editing — grounded, region-precise edits that change one element while keeping the rest of the frame intact, with layer separation, sketch completion and multi-image reference.27 cr

Topaz Generative Upscale

Topaz LabsTopaz Wonder 3.5 generative upscaler: re-renders fine detail and faces while enlarging 2× — for photos that need more than sharpening.32 cr

Topaz Image Upscale

Topaz LabsPhotoreal image upscaler with industry-leading detail recovery.16 cr

True Touch

AffogatoRefines skin texture and fine detail for photoreal portraits.10 cr

8K Upscaler

AffogatoUpscales an image to ultra-high resolution, adding fine detail tile by tile.20 cr

Alibaba Qwen Image

AlibabaStrong text rendering and editing capabilities at ~1MP default sizes.4 cr

Alibaba Qwen Image 2

AlibabaNext-gen unified generation-and-editing model with strong text rendering and realism.7 cr

Alibaba Qwen Image 2 Pro

AlibabaPremium tier with enhanced detail and composition.15 cr

Alibaba Qwen Image Edit

AlibabaEditor with strong text rendering.5 cr

Bria Background Removal

Bria AIBackground removal — commercially licensed with strong edge quality.4 cr

Bria Erase by Text

Bria AIBria FIBO-Edit Erase by Text: name the object to remove and it is gone — no mask or box needed.8 cr

Bria Eraser

Bria AIBria Eraser: removes whatever the brush mask covers and fills the gap from the surroundings.8 cr

Bria Expand

Bria AIBria FIBO-Edit 1.5 Expand: extends a photo beyond its edges to a new canvas, keeping the original pixels intact. Commercially licensed training data.8 cr

Bria Relight

Bria AIBria FIBO-Edit Relight: re-lights a photo from a chosen direction in a named look (midday, golden hour, studio, moonlight…).8 cr

ByteDance SeedEdit v3

ByteDanceInstruction-based image editing.6 cr

ByteDance Seedream 5.0 Lite

ByteDanceFast Lite version of Seedream 5.0 for high-quality text-to-image generation.7 cr

ByteDance Seedream 5.0 Lite Edit

ByteDanceLite editing endpoint with multi-image inputs.7 cr

ByteDance SeeDream v4

ByteDanceUnified image generation + editing model. High fidelity output up to 4K.6 cr

ByteDance Seedream v4 Edit

ByteDanceUnified generation + editing model with multi-image inputs (up to 10 images).6 cr

Control Light

falfal's Control Light (FLUX.2 [klein]): lifts exposure and recovers detail in low-light photos, steerable with a short description.6 cr

Fashion image

ByteDanceFashion Studio generation with resolution-aware pricing.14 cr

Finegrain Eraser

FinegrainFinegrain Eraser: draw a box around an object and it is removed, shadow and reflections included — no mask needed.36 cr

FLUX 1.1 [pro]

Black Forest LabsAn enhanced version of FLUX.1 [pro] with improved generation capabilities, delivering superior composition, detail, and artistic fidelity.8 cr

FLUX 1.1 [pro] ultra

Black Forest LabsPremium Flux 1.1 Ultra. Up to 2K output resolution across 9 aspect ratios with outstanding fidelity and composition.12 cr

Flux Rawphoto

AffogatoA candid, amateur-photo look with extra skin and texture detail.10 cr

FLUX.1 [pro] Erase

Black Forest LabsFLUX.1 [pro] Erase: removes the masked region and repaints it to match the scene.9 cr

FLUX.1 [schnell]

Black Forest LabsFast 12B-parameter Flux text-to-image model. Optimized for speed with 1-4 inference steps — ideal for rapid iteration and exploration.1 cr

FLUX.1 Kontext [max]

Text to ImageBlack Forest LabsPremium image-editing model. Requires a reference image; delivers strong typography and prompt adherence for precise edits.from 15 cr

FLUX.1 Kontext [pro]

Text to ImageBlack Forest LabsReference-image conditioned editing and restyling.from 7 cr

FLUX.1 Krea [dev]

Black Forest LabsA 12B parameter flow transformer generating high-quality images from text with exceptional aesthetics. Suitable for personal and commercial use.5 cr

FLUX.2 [dev]

EditBlack Forest LabsThe open development model in the FLUX.2 family. High-quality text-to-image generation with strong prompt adherence.from 3 cr

FLUX.2 [max]

EditBlack Forest LabsDelivers state-of-the-art image generation and advanced editing with exceptional realism, precision, and consistency.from 14 cr

FLUX.2 [pro]

EditBlack Forest LabsImage editing with FLUX.2 [pro] from Black Forest Labs. Ideal for high-quality image manipulation, style transfer, and sequential editing workflows.from 6 cr

FLUX.2 [pro] Outpaint

Black Forest LabsFLUX.2 [pro] outpainting endpoint — extend an image beyond its original frame with seamless context.9 cr

FLUX.3 Klein

Black Forest LabsBlack Forest Labs' FLUX.3 Klein — the fast, low-cost FLUX.3 tier: sub-second 1K images with FLUX.3 prompt following.2 cr

FLUX.3 Klein Edit

Black Forest LabsBlack Forest Labs' FLUX.3 Klein editing model: fast reference-guided edits with up to 10 input images.2 cr

GPT Image 2.5 Sunburst

OpenAIOpenAI's GPT Image 2.5 Sunburst — the precision variant: edits scoped tightly to the instruction, composition preserved.11 cr

GPT Image 2.5 Sunburst Edit

OpenAIOpenAI's GPT Image 2.5 Sunburst editing model: precise, subject-preserving edits.11 cr

Grok Imagine Image 2.0

xAIxAI's Grok Imagine Image 2.0: sharper detail and better prompt adherence than v1, ultrawide ratios up to 20:9.12 cr

Grok Imagine Image 2.0 Edit

xAIxAI's Grok Imagine Image 2.0 editing model: edits guided by up to 5 reference images.14 cr

HiDream O1 Image

HiDreamUnified generation, editing, and personalization up to 2K — single model handles text-to-image, editing, and custom subjects without external components.8 cr

HiDream-O1 Image Edit

HiDreamHiDream-O1 image editing endpoint. Natural-language edits up to 2K resolution.8 cr

Ideogram 3 Reframe

IdeogramIdeogram 3 Reframe: re-composes a photo onto a larger canvas, generating the new area around the centered original.12 cr

Ideogram 4.5

IdeogramIdeogram 4.5 — Ideogram's quality and typography tier: crisp text rendering, design-grade layouts, selectable quality.12 cr

Ideogram 4.5 Edit

IdeogramIdeogram 4.5 editing model: prompt-driven edits of a source image with optional masks and style references.12 cr

Ideogram Object Removal

IdeogramIdeogram Object Removal: erases the masked object and fills the background.6 cr

Ideogram Remove Background

IdeogramIdeogram's background removal — clean cutouts with strong edge fidelity. Complements BiRefNet and Bria.2 cr

Ideogram v3

IdeogramText-to-image. Best-in-class typography for posters, logos, and in-image text — ideal for branding and graphic design.12 cr

Ideogram V3 Remix

IdeogramReimagine existing images with prompt-driven variations.12 cr

Ideogram V4

IdeogramIdeogram V4 — flagship text-to-image model with best-in-class typography and prompt adherence. Pinned to the QUALITY rendering tier.5 cr

Kling Image O3

KuaishouKling Omni 3 — top-tier text-to-image with flawless consistency from the Kling team.6 cr

Kling Image V3

KuaishouText-to-image model. Cinematic aesthetic from the makers of Kling video.6 cr

Krea 2 Large

KreaKrea 2 Large — high-fidelity text-to-image with optional style reference images (up to 10) and a creativity control. Same endpoint powers Generate and Edit / Reference: with no refs it's pure text-to-image, with refs the style guides the output.12 cr

Krea 2 Medium

KreaKrea 2 Medium — faster, lower-cost Krea sibling with the same style-reference and aspect-ratio controls as Large at roughly half the credits.6 cr

Krea 3

KreaKrea 3 — Krea's third-generation image model: aesthetic, un-AI-looking photography and illustration with strong prompt adherence.5 cr

Nano Banana

GoogleGoogle's original Nano Banana (Gemini 2.5 Flash Image) — fast conversational text-to-image with strong realism. Sister model to Nano Banana Edit; lower-cost predecessor to Nano Banana 2 / Pro.8 cr

Nano Banana 2

GoogleFast, high-quality generation with thinking mode and up to 4K resolution.16 cr

Nano Banana 2 Edit

GoogleEditing — fast image editing with reference image input and up to 4K output.16 cr

Nano Banana 2 Lite

GoogleGoogle's Nano Banana 2 Lite — the efficiency tier of the Gemini image family: sub-2-second 1K images, 14 aspect ratios.9 cr

Nano Banana Lite

GoogleGoogle's Nano Banana Lite — the efficiency tier of the Gemini image family: sub-2-second 1K images, fast multi-turn edits, 14 aspect ratios.9 cr

Nano Banana Lite Edit

GoogleGoogle's Nano Banana Lite editing model: fast local edits and multi-image references at 1K.9 cr

OpenAI GPT Image 2

OpenAIExtremely detailed images with fine typography. Quality-tiered output up to 3840px.43 cr

OpenAI GPT Image 2 Edit

OpenAIEditing — fine-grained, detailed edits with multi-reference image input and quality tiers.44 cr

Qwen Image 3

AlibabaAlibaba's Qwen Image 3: dense multilingual text, infographics and layouts from long prompts.8 cr

Qwen Image 3 Edit

AlibabaAlibaba's Qwen Image 3 editing model: instruction edits with multiple reference images and faithful text.8 cr

Real-ESRGAN

Real-ESRGANFast 2x/4x upscaler for photos.2 cr

Recraft 4.1 Flash

RecraftRecraft 4.1 Flash — Recraft's fastest raster model (~1.3 s per image): clean design and illustration at the lowest price in the catalog.2 cr

Recraft v3

RecraftDesigner-focused model with realistic and vector illustration modes. Strong composition, lighting, and brand-ready aesthetics.8 cr

Recraft V4

RecraftDesigner-focused text-to-image with strong composition, lighting, and brand-ready aesthetics.8 cr

Recraft V4 Pro

RecraftPremium design-grade text-to-image for production brand workflows.50 cr

Recraft V4.1

RecraftSharper prompt control and cleaner composition. Tuned for brand systems and editorial work.8 cr

Recraft V4.1 Pro

RecraftHigh-resolution up to 2048x2048. Made for hero imagery, campaign work, and print.46 cr

Recraft V4.1 Utility

RecraftFaster, lighter variant optimized for high-volume creative workflows. Ideal for ideation, A/B exploration, and content pipelines.8 cr

Recraft V4.1 Utility Pro

RecraftHigh-resolution output with a faster, cost-efficient runtime. Designed for studios shipping large-format work at scale.46 cr

Relighting

falfal's Relighting app: applies a lighting style (studio, golden hour, neon, candlelight…) to a photo at 4K.8 cr

rembg

rembgOpen-source rembg — fast, lightweight background removal.1 cr

Seedream 5.0 Flash Edit

ByteDanceByteDance's Seedream 5.0 Flash editing model: fast multi-image edits and references, 1K and 2K output.6 cr

SoulX FlashHead

SoulXSoulX-FlashHead — unified 1.3B-parameter framework for fast head / portrait editing.3 cr

Tencent Hunyuan Image 3.0

TencentState-of-the-art text-to-image with strong visual fidelity.18 cr

Topaz Creative Upscale

Topaz LabsTopaz Bloom 2 creative upscaler: reimagines texture and detail while enlarging 2× — for illustration, concept art and heavy restoration.128 cr

xAI Grok Imagine Image

xAIHighly aesthetic image generation with wide aspect ratio support and up to 2K resolution.4 cr

xAI Grok Imagine Image Edit

xAIEditing — precise edits to existing images.5 cr

02VIDEO

Video models (71)

From Affogato, Alibaba, Black Forest Labs, ByteDance, Google, HeyGen, Kling, Kuaishou, Lightricks, Luma, MiniMax, Pika, PixVerse, Sync, Topaz Labs, VEED, Vidu, xAI. Try them in the AI Video Generator.

ModelMakerWhat it doesPrice

Affogato Video

AffogatoChoose Balanced for everyday video or Max for the highest quality.52 cr

Alibaba Wan 3.0

Image to Video · Reference to VideoAlibabaAlibaba Wan 3.0 text-to-video: native audio, smooth motion and tighter prompt adherence than Wan 2.7.from 37 cr

ByteDance SeedVR2

ByteDanceDiffusion-based video super-resolution.46 cr

FLUX.3 Edit Video

Black Forest LabsBlack Forest Labs FLUX.3 video edit: re-renders a clip under 15 s from a text instruction, preserving its motion, timing and framing.28 cr

FLUX.3 Enhance Draft

Black Forest LabsBlack Forest Labs FLUX.3 draft enhance: renders a FLUX.3 draft at full 1080p quality with synchronized audio.263 cr

FLUX.3 Extend Video

Black Forest LabsBlack Forest Labs FLUX.3 — continues an existing clip beyond its final frame, generating additional footage consistent with the original motion and scene. Input clip: MP4, under 50 MB and under 15 seconds.from 372 cr

Gemini Omni Flash Edit

GoogleGoogle Gemini Omni Flash — edits an existing video from a natural-language instruction while preserving scene coherence. Simple instructions work best; voice editing is not supported.91 cr

Gemini Omni Flash Image to Video

GoogleGoogle Gemini Omni Flash — animates a still image into video with audio, extending a single frame into coherent motion grounded in Gemini's physical understanding of how scenes behave.from 55 cr

Gemini Omni Flash Reference to Video

GoogleGoogle Gemini Omni Flash — generates video with audio from a reference image, guiding subject, motion and style from the supplied reference.from 55 cr

Gemini Omni Flash Text to Video

GoogleGoogle Gemini Omni Flash — creates video with synchronized audio from text, grounded in Gemini's real-world knowledge with improved physics understanding for coherent motion.from 55 cr

Grok Imagine Video 1.5

Image to Video · Reference to VideoxAIxAI Grok Imagine Video 1.5 text-to-video: clips up to 15 s with lip-synced speech and native audio.from 127 cr

HeyGen Lipsync

Precision · SpeedHeyGenHigh-accuracy avatar-based lip-sync with voice replacement. Slower than the Speed variant; produces better mouth shapes for close-ups.from 46 cr

Kling 3.0 4K

Image to VideoKuaishouNative 4K — professional-grade 4K video in one step, no upscaling needed. Multi-shot and native audio.from 381 cr

Kling 3.0 Pro

Image to VideoKuaishouTop-tier text-to-video with cinematic visuals, fluid motion, native audio, and multi-shot support.from 153 cr

Kling O3 4K

Reference to Video · Image to VideoKuaishouKling Omni 3 native 4K reference-to-video — composes a 4K clip from one or more reference images.from 381 cr

Kling O3 Pro

Image to Video · Video Edit · Video ReferenceKuaishouKling Omni 3 Pro — latest-generation text-to-video with cinematic visuals, multi-shot editing, and native audio.from 77 cr

Kling O3 Standard

Video Edit · Image to Video · Video ReferenceKuaishouKling Omni 3 Standard — cost-efficient latest-gen text-to-video with multi-shot and native audio.from 77 cr

LTX 2.5 Pro

Image to VideoLightricksLightricks LTX 2.5 Pro text-to-video: open-weights clips up to 10 s with native audio, from 720p up to 1080p.from 131 cr

MiniMax H3 Max

Extend · Image to VideoMiniMaxMiniMax H3 Max text-to-video: 1080p clips up to 15 s with native audio and prompt expansion.from 73 cr

MiniMax H3 Max Turbo

Extend · Image to VideoMiniMaxMiniMax H3 Max Turbo text-to-video: the fast, low-cost H3 Max tier — 1080p with native audio.from 37 cr

MiniMax H3 Reference to Video

MiniMaxMiniMax H3 reference-to-video — 2K/4K video with native audio from up to 9 subject/style reference images and up to 3 motion reference clips, each cited in the prompt by order.236 cr

Seedance 2.0

Reference to Video · Image to VideoByteDanceGenerate videos from a prompt, starting frame or up to four characters.from 220 cr

Seedance 2.5

Enhance Draft · Image to Video · Reference to Video · DraftByteDanceGenerate videos from a prompt, starting frame or up to four characters.from 160 cr

Sync v3 Lipsync

SyncFlagship lipsync with native visual intelligence. Replaces audio on a source video and re-syncs lips with high fidelity.121 cr

Topaz Video Upscale

Topaz LabsVideo upscaler. Production-grade 2x/4x upscaling.73 cr

Vidu Q4

Image to Video · Reference to VideoViduVidu Q4 preview image-to-video: animates a start frame into clips up to 16 s with native audio, 540p to 4K.from 69 cr

xAI Grok Imagine Video

Extend · Image to VideoxAIText-to-video generation with native audio.from 64 cr

Alibaba Happy Horse

Image to Video · Reference to VideoAlibaba1080p video with synchronized native audio and multilingual lip-sync.from 164 cr

Alibaba HappyHorse Video Edit

AlibabaNatural-language video editing. Supports local and global edits with optional reference images. Output preserves source aspect; capped at 15 seconds.508 cr

Alibaba Wan 2.2

A14B · Image to VideoAlibabaText-to-video. Frame-controlled output up to 720p (default 81 frames @ 16fps ≈ 5s).73 cr

Alibaba Wan 2.7

Image to VideoAlibabaLatest-generation AI video model with motion smoothness, scene fidelity, and visual coherence.from 136 cr

Alibaba Wan 3.0 Prime

Image to Video · Reference to VideoAlibabaAlibaba Wan 3.0 Prime text-to-video: native audio, smooth motion and tighter prompt adherence than Wan 2.7.from 51 cr

ByteDance Seedance 1.0 Pro

Image to VideoByteDanceSeedance Pro text-to-video. 480p/720p/1080p output with cinematic quality.from 45 cr

ByteDance Seedance 1.0 Pro Fast

Image to VideoByteDanceNext-generation video at minimal latency.from 18 cr

ByteDance Seedance 1.5 Pro

Image to VideoByteDanceHigh-quality video with audio.from 38 cr

Fashion video

ByteDanceAnimate a starting frame into a video.from 243 cr

FLUX.3 First/Last Frame to Video

Black Forest LabsBlack Forest Labs FLUX.3 — generates the video between a defined start and end frame, interpolating a smooth, coherent transition from the first image to the last.from 155 cr

FLUX.3 Image to Video

DraftBlack Forest LabsBlack Forest Labs FLUX.3 — animates a single still image into video, extending one frame into coherent, natural motion.from 55 cr

FLUX.3 Text to Video

DraftBlack Forest LabsBlack Forest Labs FLUX.3 — generates video directly from a text prompt, translating a written description into motion, composition and scene. Audio is generated with the video.from 55 cr

Grok Imagine Video 1.5 Lite

Image to VideoxAIxAI Grok Imagine Video 1.5 Lite text-to-video — the cheap tier: clips up to 15 s with lip-synced speech and native audio.from 28 cr

Kling 1.6 Pro

Image to VideoKuaishouText-to-video. Professional-tier video generation with strong motion fidelity.from 87 cr

Kling 1.6 Standard

Image to VideoKuaishouCost-efficient text-to-video for fast iteration.from 51 cr

Kling 2.1 Master

Image to VideoKuaishouPremium Kling 2.1 Master text-to-video. Cinematic motion with precise prompt adherence.from 254 cr

Kling 2.5 Turbo Pro

Image to VideoKuaishouFast cinematic text-to-video.from 64 cr

Kling 2.6 Pro

Image to VideoKuaishouTop-tier text-to-video with cinematic visuals and native audio.from 127 cr

Kling 3.0 Standard

Image to VideoKuaishouCinematic text-to-video with multi-shot support.from 115 cr

Kling 3.0 Turbo Pro

Image to VideoKuaishouKling 3.0 Turbo Pro text-to-video: fast Kling 3.0 at the pro tier.from 127 cr

Kling 3.0 Turbo Standard

Image to VideoKuaishouKling 3.0 Turbo Standard text-to-video: the fast Kling 3.0 tier.from 102 cr

Kling AI Avatar v2

KlingNatural lip-sync and head motion from a photo and a voice clip — the volume engine, at roughly a third of the premium price.102 cr

Kling O1

Video ReferenceKlingGenerate new footage guided by a video reference, images and characters.from 92 cr

LTX 2.3 22B Distilled — Reference V2V

LightricksGenerates video with audio from a reference video + prompt; supports start/end image guidance and camera LoRA.25 cr

LTX 2.5 Fast

Image to VideoLightricksLightricks LTX 2.5 Fast text-to-video: open-weights clips up to 20 s with native audio, from 720p up to 4K.from 98 cr

LTX Video 13B Distilled

LightricksFrame-controlled generation (default 121 frames @ 24fps ≈ 5s).19 cr

LTX Video 2.3 Fast

Image to VideoLightricksLower latency video generation.from 55 cr

LTX Video 2.3 Pro

Image to VideoLightricksHigh-quality fast AI video generation with audio.from 73 cr

Luma Dream Machine Ray 2

Image to VideoLumaPhotoreal text-to-video. 540p/720p/1080p resolutions with 5s and 9s durations.from 55 cr

MiniMax Hailuo 02 Pro

Image to VideoMiniMaxText-to-video at 1080p.from 87 cr

MiniMax Hailuo 02 Standard

Image to VideoMiniMaxText-to-video at 768p.from 49 cr

MiniMax Hailuo 2.3 Pro

Image to VideoMiniMaxText-to-video at 1080p.89 cr

MiniMax Hailuo 2.3 Standard

Image to VideoMiniMaxText-to-video at 768p.from 51 cr

OmniHuman 1.5

ByteDanceThe premium performance engine — expressive gestures and motion you can direct with a text prompt.290 cr

OmniHuman 2

ByteDanceByteDance OmniHuman 2 — the premium performance engine, now cheaper per second: expressive gestures and camera you can direct with a text prompt.191 cr

Pika v2.2

Image to VideoPikaText-to-video — dynamic 1080p generations with sharp details and cinematic camera moves.from 37 cr

PixVerse C1

Image to VideoPixVerseFilm-grade text-to-video with native audio, up to 1080p and 15 seconds.from 46 cr

PixVerse V6

Image to VideoPixVerseLatest generation text-to-video with audio, up to 1080p and 15 seconds.from 41 cr

Seedance 2.0 Fast

Image to Video · Reference to VideoByteDanceGenerate videos from a prompt, starting frame or up to four characters.from 176 cr

Seedance 2.5 Reference

DraftByteDanceGenerate a new video guided by image or video references and characters.from 160 cr

Sync Lipsync 2

DraftSyncSync lipsync-2 draft: a quick, low-cost lipsync preview at about a third of the sync-3 price — enhance it with sync-3 when the take is right.46 cr

VEED Subtitles

VEEDVEED's Subtitles API — transforms raw footage into polished, published-ready video with on-screen subtitles.19 cr

Videozypher

AffogatoRecasts a reference video with your own character, keeping the motion.52 cr

Vidu Q3

Image to VideoViduLatest text-to-video with native audio, up to 1080p.from 140 cr

03AUDIO

Voice and audio models (12)

From CassetteAI, ElevenLabs, Google, Kokoro, MiniMax, Stability AI. Try them in the AI Voice Generator.

ModelMakerWhat it doesPrice

CassetteAI

CassetteAINear-instant stereo tracks at draft-tier pricing.from 4 cr

CassetteAI SFX

CassetteAIFast, flat-price sound effects up to 30 seconds.2 cr

ElevenLabs Music

ElevenLabsFull songs with vocals, trained exclusively on licensed data — the commercial-safe lane for client work.from 120 cr

ElevenLabs SFX v2

ElevenLabsCinematic sound effects from a short description.2 cr

ElevenLabs Turbo v2.5

ElevenLabsFast, half-price ElevenLabs voices in 32 languages — great for drafts and iteration.10 cr

ElevenLabs v3

ElevenLabsElevenLabs' most expressive voices — write your script in any of 70+ languages (auto-detected), with inline audio tags like [whispers] and [laughs].20 cr

Kokoro

KokoroLightweight English-only voices at draft-tier pricing.4 cr

Lyria 3

GoogleGoogle's music model — quick tracks from a prompt.8 cr

Lyria 3 Pro

GoogleGoogle's flagship music model — vocals, structure, metadata.16 cr

MiniMax Music 3

MiniMaxFull songs with vocals from a style prompt and your lyrics.from 12 cr

MiniMax Speech 2.8 HD

MiniMaxHigh-fidelity voices across 37 languages, with a wide character range.20 cr

Stable Audio 3

Stability AILicensed-data instrumentals and soundscapes up to six minutes.8 cr

FAQQUESTIONS

How much does each model cost?

Every generation costs credits, and the figure next to each model is the lowest a single run of it costs. Some models price by length or resolution, shown as “from”. Plans come with monthly credits; see the pricing page.

Can I call these models from code?

Yes. Every model listed here is also served by the Affogato API and MCP server, at the same credit price as in the studios.

How current is this list?

It is read from the live catalog, so a model appears here within minutes of being switched on and disappears when it is retired.