Beta
Docs
Providers & Models

Image Providers

All image generation and editing models currently available in Hylope, organized by provider.

Picking an image model comes down to what you're optimizing for: raw prompt-following, precise multi-reference editing, output resolution, or speed. The tables below break down every model by provider so you can match one to the job, whether you're working in Image Generation, Image Editing, or a related create mode (Previz, Product Swap, Avatar where listed).

Not sure where to start?

Nano Banana 2 (Google Gemini) is the default across most modes — a solid all-around balance of quality, speed, and cost. Reach for a specialist below when you need something it isn't built for, like Nano Banana Pro's higher input-reference count or Seedream's native high-resolution output.

Every model is available on every paid plan — billing is per generation from your credit balance. Model coverage matches the in-app Image Gen / Edit pickers. For pricing, see the Pricing page.


Google Google

Gemini (Nano Banana)

ModelModesNotes
Nano Banana 2Gen, Edit, Previz, Swap, AvatarDefault balance of quality/speed/cost; up to 10 inputs
Nano Banana LiteGen, Edit, Previz, Swap, AvatarFastest/cheapest; 1K only
Nano Banana ProGen, Edit, Previz, Swap, AvatarThinking + Search grounding; up to 14 inputs; 1K, 2K, or 4K
Nano BananaGen, Edit, Swap, AvatarPrevious gen; 1K only; up to 3 inputs

Unique: Search grounding, thinking mode (Pro).

Imagen

No longer available for new generations

Google shut down Imagen 4 — you can't start a new job with it. Past Imagen generations still restore fine. For new work, use Gemini, Qwen, Seedream, Flux, or OpenAI instead.

ModelNotes
Imagen 4 Fast / 4 / UltraLegacy photoreal diffusion

Alibaba Alibaba

Qwen

ModelModesNotes
Qwen 2.0Gen, Edit, Swap, AvatarBalanced default
Qwen 2.0 ProGen, Edit, Swap, AvatarStronger text rendering & realism
Qwen MaxGenMaximum quality
Qwen PlusGenBalanced budget option
Qwen Edit MaxEdit, Swap, AvatarEdit-focused

Unique: Style presets, negative prompts, seeds · 1K–2K

Wan Image

ModelModesNotes
Wan 2.7 ImageGen, Edit, Previz, Swap, AvatarUp to 9 refs; 1K/2K
Wan 2.7 Image ProGen, Edit, Previz, Swap, Avatar4K T2I; 2K edit
Wan 2.6 T2I / ImageGen / EditLegacy compatibility
Wan 2.7 Video EditPrevizStylized motion transfer

BytePlus BytePlus

Seedream

ModelModesNotes
Seedream 5.0 ProGen, Edit, Avatar, Previz, SwapPrecise multi-ref editing; 1K–2K
Seedream 5.0 LiteGen, EditDefault; native 2K–3K
Seedream 4.5Gen, EditPrevious generation

Unique: Strong photorealism; high native resolutions on Lite.


Black Forest Labs Black Forest Labs

Flux 2

ModelNotes
Flux 2 Klein 4BFast, affordable default
Flux 2 Klein 9BBetter quality lightweight
Flux 2 ProProfessional prompt following
Flux 2 FlexReference image support
Flux 2 MaxHighest detail

Available in Gen, Edit, Previz, Swap, Avatar · 1K–2K · seeds · up to 8 refs (4 on Klein)


OpenAI

ModelModesNotes
ChatGPT Image 2.0Gen, Edit, Previz, Swap, AvatarHigh-fidelity refs; up to 16 inputs; 1K–4K (4K on wide ratios)

Microsoft

ModelModesNotes
MAI Image 2.5 FlashGen, EditFast 1K
MAI Image 2.5Gen, EditHigher quality 1K

Ideogram

ModelModesNotes
Ideogram 4.0Gen, Edit2K T2I + remix; rendering speed control

Krea

ModelModesNotes
Krea 2 LargeGenPhotorealism, textured looks
Krea 2 MediumGenIllustration / anime / expressive
Krea 2 Medium TurboGenFaster medium variant

xAI xAI (Grok)

ModelModesNotes
Grok ImagineGen, Edit, Previz, Swap, AvatarFast
Grok Imagine QualityGen, Edit, Previz, Swap, AvatarHigher quality

Up to 3 input images · 1K–2K


Kling (image)

ModelModesNotes
Kling 3.0GenStandard 1K/2K T2I
Kling 3.0 OmniGen, Edit, Previz, Swap, AvatarPrompt-addressable refs; up to 4K

Recraft Recraft

ModelNotes
Recraft V4.1 / V4.1 ProCurrent design-focused line
Recraft V4.1 Utility / Utility ProUtility variants
Recraft V4 / V4 ProPrevious generation

Optimized for graphic design and illustration.


Reve Reve

ModelModes
ReveImage Gen
Reve FastImage Gen
Reve EditImage Edit
Reve Edit FastImage Edit

Very fast generation and editing.


Quality tiers

  • 1K — fastest / baseline
  • 2K — common balanced tier
  • 3K / 4K — where the model supports it (e.g. Seedream Lite at 3K; Nano Banana Pro, Wan Pro, OpenAI, and Kling Omni at 4K)

On this page