Every model, one typed interface

110+ curated nodes: image, video, language and audio models, plus media utilities. Each one carries typed ports, real defaults and a pinned version.

Read the run-a-node guide

The node catalog

All 117Generate Video 18Generate Images 34Edit & Analyze Image 32Upscale Image 5Generate Audio 8Transcribe Audio 2Language 113D 1Media utilities 6

Generate Images

Text and reference images into a new still.

Nano Banana Pro output
a luxury perfume bottle on black marble, dramatic rim lighting, campaign photography

Nano Banana Pro

Conversational image generation and editing with Nano Banana Pro

GPT Image 2.5 Sunburst output
a hot air balloon festival at sunrise, wide landscape, rich saturated color

GPT Image 2.5 Sunburst

OpenAI's highest-quality image generation and editing tier, up to 4K

Seedream 5.0 Pro output
a marble sculpture of a dancer mid-turn, museum lighting, ultra-detailed

Seedream 5.0 Pro

Seedream 5.0 text-to-image with reference re-mixing and strong typography

Flux 2 Pro output
a glass terrarium city under a desert sky, intricate detail, cinematic lighting

Flux 2 Pro

FLUX.2 text-to-image up to 4MP, optionally guided by a reference image

Ideogram v4 Quality output
an enamel pin design that reads "TRAIL CREW", crisp lettering on forest green

Ideogram v4 Quality

Ideogram 4.0 at its highest-fidelity rung

Recraft v4.1 output
a flat vector illustration of a mountain cable car, muted alpine palette

Recraft v4.1

Design-led text-to-image with precise prompt adherence.

Grok Imagine 2 output
a hummingbird hovering at a trumpet flower, high-speed photography, crisp detail

Grok Imagine 2

Grok Imagine 2.0 text-to-image

Muse Image output
an oil painting of a harbour at dusk, thick impasto brushwork, lantern reflections

Muse Image

Meta's image generation and editing model, blending a reference into the prompt

Flux 1.1 Pro output
a lighthouse in a storm, spray lit by the beam, dramatic photographic realism

Flux 1.1 Pro

Top-tier FLUX text-to-image with strong prompt adherence

Flux 2 Klein output
a fox curled asleep on a mossy log, dappled forest light, photorealistic

Flux 2 Klein

Sub-second FLUX.2 generation for drafts and latency-critical steps

Flux Dev output
an astronaut resting in a hammock between two palm trees on an alien beach, golden hour, cinematic

Flux Dev

High-quality text-to-image with FLUX.1 [dev]

Flux Schnell output
a red panda barista pulling an espresso shot, warm morning light, shallow depth of field

Flux Schnell

v2.0.0

Fast, cheap text-to-image with FLUX.1 [schnell]

GPT Image 1 Mini output
a friendly robot watering plants on a sunny balcony, clean digital illustration

GPT Image 1 Mini

Text-to-image with GPT Image 1 Mini, optionally guided by a reference image

GPT Image 1.5 output
a ceramic teapot surrounded by wildflowers, editorial product photography

GPT Image 1.5

Text-to-image with GPT Image 1.5, optionally guided by a reference image

GPT Image 2 output
a cozy reading nook beside a rain-streaked window, warm lamp light, watercolor style

GPT Image 2

Text-to-image with GPT Image 2, optionally guided by a reference image

GPT Image 2.5 Flare output
a bowl of ramen photographed from above, steam curling, moody restaurant light

GPT Image 2.5 Flare

OpenAI's fast image generation tier, up to 4K

Ideogram v3 Balanced output
a bold gig poster that reads "NIGHT MARKET" over neon street food stalls

Ideogram v3 Balanced

Ideogram v3 at the middle speed/quality rung

Ideogram v3 Quality output
an art-deco book cover that reads "THE GILDED HOUR", gold foil on deep teal

Ideogram v3 Quality

Highest-fidelity Ideogram v3 text-to-image

Ideogram v3 Turbo output
a vintage travel poster that reads "KYOTO" with cherry blossoms and a pagoda, bold clean typography

Ideogram v3 Turbo

Fast text-to-image with strong, legible text rendering

Ideogram v4 Balanced output
a festival poster that reads "HARVEST MOON" over paper-cut hills

Ideogram v4 Balanced

Ideogram 4.0 at the middle speed/quality rung

Krea 2 Medium output
an anime-style rooftop at dusk, city lights blooming behind a lone figure

Krea 2 Medium

Expressive illustration and anime styles.

Nano Banana output
a stack of vintage suitcases in a sunlit attic, dust motes in the air

Nano Banana

Conversational image generation and editing with Nano Banana

Nano Banana 2 output
a corgi puppy wearing a tiny chef's hat, studio product photography, soft light

Nano Banana 2

Conversational image generation and editing with Nano Banana 2

Nano Banana 2 Lite output
a paper crane resting on a windowsill beside a cup of tea, soft daylight

Nano Banana 2 Lite

Conversational image generation and editing with Nano Banana 2 Lite

Pruna Image (Fast) output
a letterpress card that reads "GOOD MORNING", warm cream stock

Pruna Image (Fast)

Sub-second text-to-image with strong text rendering

Recraft v3 output
a minimalist flat illustration of a steaming coffee cup, soft pastel palette

Recraft v3

High-quality text-to-image with style control

Retro Diffusion Animation (Pixel Art) output
a knight in silver armor — four-angle walk cycle

Retro Diffusion Animation (Pixel Art)

Pixel-art animation spritesheet generation (walk/idle cycles, 4-direction characters)

Retro Diffusion Fast (Pixel Art) output
a brass treasure chest overflowing with gold coins

Retro Diffusion Fast (Pixel Art)

Fast pixel-art / retro game-asset generation

Retro Diffusion Plus (Pixel Art) output
an ancient spellbook with glowing runes on the cover

Retro Diffusion Plus (Pixel Art)

High-quality pixel-art generation

Retro Diffusion Tile (Pixel Art) output
mossy cobblestone dungeon floor

Retro Diffusion Tile (Pixel Art)

Pixel-art tilesets, single tiles and map objects

SDXL output
a majestic snow leopard perched on a rocky cliff at dawn, ultra-detailed, photorealistic

SDXL

Text-to-image with Stable Diffusion XL

Seedream 4.5 output
a weathered brass diving helmet on a workshop bench, shafts of dusty light

Seedream 4.5

Generate or re-mix from references, up to 4K.

Seedream 5.0 Lite output
a ceramic bowl of ripe figs on linen, natural window light, still life

Seedream 5.0 Lite

Seedream 5.0's cheaper tier, with the same reference re-mixing

Z-Image Turbo output
a neon-lit ramen stall in the rain, reflections on wet asphalt, cinematic

Z-Image Turbo

Six-billion-parameter text-to-image in a handful of steps

Run one from your own code

Open the studio