Meta’s Muse Spark Open now available in Amazon SageMaker JumpStart
Machine Learning Blog
This article announces the availability of Meta's Muse Spark Open, a coding-optimized open-weight model, in Amazon SageMaker JumpStart for easy deployment on AWS endpoints.
- Muse Spark Open features 816B total parameters with 42B active per token and supports 1M token context window
- Trained on real developer workflows including multi-file refactors, debugging, and long-running coding tasks
- Available in two variants: BF16 and NVFP4 quantized, running on ml.p6-b300.48xlarge and ml.p5e.48xlarge instances respectively
- Supports multimodal inputs (text, image, document, audio, video) and returns reasoning alongside answers
- Deploy via SageMaker Studio or Python SDK without managing serving infrastructure
- Ideal for coding assistants over private repositories, long-running agents, and multimodal engineering workflows
Muse Spark Open enables developers to deploy a powerful reasoning model for code-focused tasks while maintaining control over infrastructure and data privacy in their AWS account.
The AWS News Feed is currently looking for gold sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.
The AWS News Feed is currently looking for silver sponsors. If you want to support the AWS community and reach a large audience of AWS professionals, consider sponsoring the AWS News Feed.