Nodes

ElevenLabs' most expressive voices for narration and characters. Up to 1,000 characters per call, one flat price. Returns an MP3.

Nine natural voices in 10 languages, steerable with a plain-language style instruction (tone, pace, emotion). Up to 1,000 characters per call.

Natural speech in 15+ languages with low latency. Up to 1,000 characters per call, one flat price. Returns an MP3 (or WAV/OGG/FLAC).

A full song with vocals from your lyrics and a style description. One flat price per song. Returns an MP3.

Cut the subject out of any image and get a clean transparent PNG. Hair and soft edges kept.

Upscale an image while sharpening edges and fine detail, without the plastic look. One flat price.

Up to 4-megapixel images with FLUX 1.1 Pro Ultra, plus a raw mode for natural, less processed photography.

Black Forest Labs' workhorse pro model: excellent image quality, prompt adherence and variety, fast.

Black Forest Labs' latest image model: generate from a prompt or edit with up to 10 reference images, from 768px to 4K.

The popular 12B open-weight FLUX model: strong quality at a low price, with optional image-to-image.

The premium Kontext editor: maximum quality text-guided edits and the best typography of the FLUX family.

Edit an image with a text instruction while keeping the subject consistent. One flat price per edit.

The classic fast upscaler: enlarge an image 2x or 4x, with optional face enhancement. The cheapest way to add resolution.

Turn a text prompt into one 1-megapixel image in about two seconds. Returns a WebP image URL.

OpenAI's GPT Image 1.5: better instruction following, transparent backgrounds and edits from reference images.

OpenAI's image model with close instruction following, sharp text rendering and detailed edits from reference images.

xAI's image model: vivid, photorealistic images and edits from a prompt, at a low flat price.

Ideogram's newest model, known for precise typography and layouts. Text-to-image; pick low, medium or high quality.

The fastest, cheapest Ideogram v3: stunning realism, creative designs and reliable text in images.

Edit one or more images with a plain-language instruction in about a second: swap backgrounds, relight, combine subjects.

A sharp image from a text prompt in under a second. The cheapest way for an agent to make a picture.

Google's fast image model: crisp text in images, multi-image editing and up to 4K output. Pass reference images to edit.

Google's highest-quality image model: studio-grade detail, accurate text and multi-image composition, at 1K, 2K or 4K.

The original Nano Banana from Google: fast generation and conversational editing that keeps faces and products consistent.

Repair scratched, faded or damaged photos: removes damage, sharpens and restores natural colour.

Multi-image editing from Alibaba's Qwen: combine people, products and scenes, or change text in images precisely.

Image generation with a designer's eye: clean compositions, brand-friendly styles, one flat price.

ByteDance's upgraded image model with stronger spatial understanding and world knowledge, at 2K or 4K for one price.

ByteDance's image model with built-in reasoning and example-based editing, at 2K or 3K, one flat price.

A super-fast 6B text-to-image model with excellent photorealism and text, at 1 megapixel for a fraction of a cent.

Validate a 17-character VIN and return make, model, year, plant and body details plus an ISO 3779 breakdown.

Run one Google search and get the first page of organic results as JSON: title, URL, description and position, plus related queries.

Fast 5-second videos with sound from a text prompt or a starting image, at 720p or 1080p.

ByteDance's Seedance Lite: 5-second videos from text or an image, with cinematic camera moves, at 480p or 720p.

Animate a still image into a 5-second video clip with Wan 2.2: fast and cheap, at 480p or 720p.