Mistral AI’s Pixtral Large Model is Now Available as a Fully Managed, Serverless Offering within Amazon Bedrock
AWS is announcing that the Pixtral Large 25.02 model—Mistral AI’s frontier multimodal model—is now available as a fully managed, serverless offering within Amazon Bedrock. As the first major cloud provider to deliver Pixtral Large as a fully managed, serverless model, this release boosts the availability of Mistral AI’s advanced image understanding with powerful text processing capabilities.
The Pixtral Large model shines in document analysis, visual reasoning, and natural image comprehension while simultaneously maintaining the text capabilities delivered from Mistral Large 2. The model empowers organizations to process complex visual-text tasks with both accuracy and efficiency, boasting a 128K context window optimized for low latency.
A key power of the Pixtral Large model is its extensive language support, offering dozens of languages—including English, French, German, Spanish, Italian, Chinese, Japanese, Korean, Portuguese, Dutch, and Polish. Outside of its global applicability, the Pixtral Large model is trained on over 80 programming languages, including Python, Java, C, C++, JavaScript, Bash, Swift, and Fortran, forwarding robust code generation and interpretation.
Like other Mistral AI models, Pixtral Large 25.02 is transparent and customizable, making it suitable for industries with more rigorous compliance and regulatory requirements. With an agent-centric design, built-in function calling, and JSON output formatting, Pixtral Large 25.02 seamlessly integrates with existing systems.
With its newfound availability on Amazon Bedrock, organizations can utilize Pixtral Large 25.02 without having to deal with headaches of infrastructure provisioning or management. Enterprises can scale usage based on actual demand, without any upfront commitments or capacity planning—you only pay for what you use.
To learn more about Pixtral Large 25.02 and its AWS availability, please visit https://aws.amazon.com/ or https://mistral.ai/.