Amazon Bedrock Nova Canvas
Free使用 Amazon Nova Canvas 模型进行图像生成。
FreeFree tier
Inputs: textOutputs: image
About Amazon Bedrock Nova Canvas
A Model Context Protocol (MCP) server that integrates with Amazon Bedrock's Nova Canvas model for AI image generation. It allows users to generate high-quality images from text descriptions with advanced controls such as negative prompts, configurable dimensions and quality settings, seed-based reproducibility, and batch generation. The server requires an active AWS account with Amazon Bedrock access and Node.js 18+.
Key Features
High-quality image generation from text descriptions using Amazon Nova Canvas
Advanced control through negative prompts to refine composition
Flexible configuration options for image dimensions and quality (standard/premium)
Deterministic image generation with seed control for reproducibility
Batch generation (1-5 images) with configurable prompt adherence (cfg_scale)
Robust input validation and error handling
Pros & Cons
Pros
- Leverages Amazon's advanced Nova Canvas model for high-quality results
- Offers fine-grained control over generation parameters (seed, cfg_scale, dimensions, quality)
- Supports negative prompts to exclude unwanted content
- Enables deterministic outputs via seed parameter, useful for testing and iteration
- Simple MCP integration with popular AI assistants like Claude Desktop
Cons
- Requires an active AWS account with Bedrock access and properly configured credentials
- Costs associated with AWS Bedrock usage (server itself is free and open source)
- Limited to the Nova Canvas model; no support for other image generation models
- Requires Node.js 18 or later for local installation
Best For
Generate images from detailed text descriptionsCreate image variations by using negative prompts to exclude elementsProduce multiple image options in a single batch for quick iterationIntegrate AI image generation into MCP-compatible client applications like Claude Desktop
FAQ
What is the generate_image tool?
It is the primary tool of this MCP server that creates images from text descriptions using Amazon Bedrock's Nova Canvas model.
How do I configure AWS credentials?
You can set credentials via environment variables (AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, AWS_REGION), an AWS credentials file, an AWS profile environment variable, or an IAM role when deployed on AWS infrastructure.
What parameters are available for image generation?
The generate_image tool accepts a required prompt (1-1024 characters), optional negativePrompt, width and height (default 1024), quality (standard/premium), cfg_scale (1.1-10, default 6.5), seed (0-858993459, default 12), and numberOfImages (1-5, default 1).
Is this MCP server free to use?
The server itself is open source and free. However, using it will incur charges from AWS for Amazon Bedrock usage.