Generative AI
Overview Our Story Customers Partners

Build creative applications with HAQM Nova

State-of-the-art image and video generation models

Try HAQM Nova
Build with HAQM Nova

HAQM Nova Creative Models

HAQM Nova creative models include HAQM Nova Canvas for image generation and HAQM Nova Reel for video generation. These models are designed to generate high-quality images and videos from text and image inputs, providing customizable visual content for various applications. Through seamless integration with HAQM Bedrock, developers can build and scale generative AI applications with HAQM Nova creative models.

Key capabilities

HAQM Nova Canvas and HAQM Nova Reel allow multiple ways for you to control and customize the generated content. You can provide a reference image to guide your image or video generation. HAQM Nova Canvas also lets you use text inputs to determine the color scheme or “color palette” for the image. With camera motion control in HAQM Nova Reel, you can guide camera motion with text prompts like “zoom” and “dolly forward” to get the exact visual needed for your video.

With HAQM Nova Canvas, customers can fine-tune the model on their proprietary data to generate fully customized images that align with to specific requirements and style guidelines.

HAQM Nova Canvas allows precise image editing operations where you can replace objects or backgrounds in an input image using text prompts. You can also easily remove a background with the background removal feature, leaving the subject of the image unchanged.

With HAQM Nova Reel 1.1, users can generate multi-shot videos up to 2-minutes in length and style consistency across shots. Multi-shot videos offer two modes: automated mode (single prompt and specifying total video duration) and manual mode (multiple prompts and an optional image for each 6-second shot).

HAQM Nova Reel and HAQM Nova Canvas have built-in controls for the safe and responsible use of AI. These include watermarking for traceability, content moderation, and indemnification.

HAQM Nova Canvas

HAQM Nova Canvas is an image generation model that creates professional grade images from text and image inputs. HAQM Nova Canvas is ideal for a wide range of applications such as advertising, marketing, and entertainment.

Example images generated by HAQM Nova Canvas

“A cactus with googly eyes holding a magic wand emitting sparks”

“Hand drawn pencil illustrated character design of a shape shifting robot in a sci-fi city”

“A pink hot air balloon”

“An octopus at the bottom of the ocean wearing a party hat”

“A bright running shoe”

“Two cartoon dogs hugging each other”

“A dinosaur sitting in a teacup”

“A very fancy French restaurant”

“Black-and-white photo, character study, multi-angle”

HAQM Nova Reel

HAQM Nova Reel is a video generation model that supports the generation of short videos from input text and images. HAQM Nova Reel provides camera motion controls using natural language inputs.

Introducing HAQM Nova Reel 1.1

HAQM Nova Reel 1.1 enables the generation of multi-shot videos up to 2-minutes in length and style consistency across shots. In addition, HAQM Nova Reel 1.1 provides quality and latency improvements in 6-second single-shot video generation over HAQM Nova Reel 1.0.

Videos generated by HAQM Nova Reel 1.1

Underwater Raccoon

This video was generated with HAQM Nova Reel 1.1 in automated mode, with a single, simple prompt: “A cute raccoon playing guitar underwater.”

Color Burst

Using automated mode, this video was generated with details to provide guidance and control. The prompt included "explosion of colored powder against black background. Start with slow-motion closeup of single purple powder burst. Dolly out revealing multiple powder clouds in vibrant hues colliding mid-air."

Above Zero

Using manual mode, this video used individual images per shot generated by HAQM Nova Canvas. The images helped to refined thematic set of prompts and consistency. The prompts included "dramatic overcast day at a massive glacier terminus, moody cinematic."

Animal Sanctuary

Using manual mode, this video included 10 specific shots across an animal santuary. The prompt for the first shot is, "Silhouette of woman with dark curly hair reflected in glass elevator, breathing exercise, slight fog on glass surface, cool ambient lighting"

Lone Wolf

Using automated mode, this video captures character consistency of a wolf, with distinctive features: silver-tipped charcoal fur, piercing amber eyes, and a small white patch on its chest. The prompts provided recommended minimal movement, focusing on composition and atmosphere.

See extended HAQM Nova Reel gallery