Source URL: https://aws.amazon.com/blogs/aws/openai-open-weight-models-now-available-on-aws/
Source: AWS News Blog
Title: OpenAI open weight models now available on AWS
Feedly Summary: AWS continues to expand access to the most advanced foundation models with OpenAI open weight models now available in Amazon Bedrock and Amazon SageMaker JumpStart. Accessing these new models from OpenAI on AWS, gpt-oss-120b and gpt-oss-20b, gives you more freedom to innovate and choose the right model for your specific use cases while maintaining complete control over your infrastructure and data.
AI Summary and Description: Yes
**Summary:** AWS has introduced two new OpenAI models, gpt-oss-120b and gpt-oss-20b, available through Amazon Bedrock and SageMaker JumpStart, enabling developers to harness advanced AI capabilities with customizable reasoning and extensive usability options. These offerings enhance the ability to build AI applications while maintaining control over infrastructure and data, significant for professionals working in AI and cloud security domains.
**Detailed Description:**
AWS is making strides in the integration of advanced AI models, specifically OpenAI’s new models, which are designed to expand businesses’ AI capabilities while ensuring control over their operational infrastructure.
Key insights include:
– **Advanced Foundation Models:**
– Introduction of gpt-oss-120b and gpt-oss-20b models for text generation and reasoning tasks.
– These models are noted for their coding, scientific analysis, and mathematical reasoning capabilities.
– **Customizability and Control:**
– Both models come with adjustable reasoning levels (low/medium/high) tailored to use case requirements.
– They offer complete control over the associated data and infrastructure via services like Amazon Bedrock and SageMaker JumpStart.
– **Innovative Features:**
– A maximum context window of 128K enhances the models’ viability for complex tasks.
– External tools can be integrated to amplify the models’ functionalities.
– **Operational Effectiveness:**
– The combination of Amazon Bedrock and SageMaker allows users to freely experiment with various models and capabilities without heavy re-coding, representing a flexible and strategic advantage in the AI landscape.
– **Accessibility and Usability:**
– Users can easily request access to these models, test them via a user-friendly console, and deploy them in production using AWS SDKs/CLI.
– **Deployment Guidelines:**
– A streamlined process for deploying models in production using the Amazon SageMaker interface and Bedrock’s serverless features for easy operationalization.
– **Security and Compliance Focus:**
– Safety measures are emphasized, highlighting that models are developed with security in mind and go through rigorous evaluations.
– **Implementation Examples:**
– Sample code snippets demonstrate practical applications and illustrate the ease of building AI agents with the new models.
– **Additional Resources:**
– Links for deeper insights into pricing and detailed model parameters are provided to guide operational planning.
These developments signify AWS’s commitment to leveraging AI’s transformative potential while ensuring that clients are equipped with the tools and security measures necessary to navigate compliance and operational standards in an increasingly data-driven environment.