Families
Browse model families.
Dreamina
FLUX
Black Forest Labs' FLUX models: FLUX 3 for top-quality generation and multi-reference editing, and FLUX.2 [klein] for near-instant drafts and edits.
Google's AI lineup in our catalog spans the Nano Banana image family (Nano Banana, Nano Banana 2, Nano Banana Pro and the Edit Fast / Edit Ultra variants) plus Gemini 2.5 Flash for multimodal LLM work. Nano Banana Pro is the strongest reasoning tier of the line; Nano Banana Pro Edit Ultra is our reference image-edit specialist.
Grok
Hailuo
Happy Horse
Happy Horse is Alibaba's text- and image-to-video model family. One member today: Happy Horse 1.0, which generates a clip from a written prompt or animates a single still image into a short video. Useful as a fast text-to-video starting point and as a lighter alternative to the Seedance and Wan i2v flagships for high-iteration cycles.
Hunyuan
Ideogram
Ideogram's image models are built around typography: legible lettering, logos, posters and graphic layouts, plus editing and inpainting of existing images.
Kling
Krea
Krea's image models follow plain-language instructions: fast edits that change only what you ask for, multi-image composites and text-to-image generation.
Minimax
OpenAI
OpenAI's image lineup in our catalog: GPT Image 1.5 (the only model in the family that supports native transparent backgrounds) and GPT Image 2 (the latest flagship — strong instruction following, sharp text rendering, precise edits with reference images). Both excel at typography and structured layouts.
Pruna
Pruna AI builds speed-optimized variants of leading open-source AI models — same capability, sub-second inference, lower cost. Our Pruna catalog spans Pruna Video (fast text/image-to-video), Pruna P-Video Avatar (the fastest talking-avatar / lip-sync model on the market), Pruna P-Image-Upscale (sub-second upscaling up to 128 MP) and Pruna P-Image-Edit (twenty-task multi-image editor).
Seedance
Seedance is ByteDance's video lineage — cinematic motion, native audio, multimodal reference inputs. The family row is wired to Seedance 2.0 Video Edit today; the sibling Seedance 2.0 Text-to-Video model is in the catalog but its family_id isn't set yet (admin can pin it to this family to make the lineage discoverable on the family page).
Seedream
Seedream is ByteDance's image-generation lineage — tuned for crisp typography, strong prompt adherence and clean commercial compositions. Our catalog covers Seedream 4.5 (text-to-image base), Seedream 4.5 Edit (image editing with reference images and written instructions) and Seedream 5 Lite (the faster generation tier of the latest release).
Wan 2.2
Wan 2.2 is the previous-generation Alibaba video lineup, kept in the catalog for its speed-tier i2v variants. Wan 2.2 i2v Fast is the PrunaAI-optimised image-to-video model engineered for short turnaround; Wan 2.2 i2v 480p Ultra-Fast adds LoRA support at fixed 480p — useful for brand-styled iteration at high volume. For new commercial work, Wan 2.7 I2V is usually the better choice.