Amazon Bedrock Custom Model Import now supports Qwen models

Amazon Bedrock Custom Model Import now supports Qwen models. You can now import custom weights for Qwen-based architectures, including models like Qwen 2.5 Coder, Qwen 2.5 VL, and QwQ 32B. This enables you to bring your own customized Qwen models into Bedrock and deploy them in a fully managed, serverless environment—without having to manage infrastructure or model serving.

Qwen models are known for their strong performance across a wide range of modalities and tasks. Qwen 2.5 Coder is optimized for code generation and understanding, making it well-suited for tasks like code completion, bug fixing, and code translation. Qwen 2.5 VL is a multimodal model capable of both text and vision tasks such as visual question answering, image captioning, and document analysis. QwQ 32B, excels in complex reasoning tasks and achieving performance comparable to larger models while being more efficient.

You can get started by importing your custom Qwen model in the custom models page of the Amazon Bedrock console or by referring to this guide. To see what all architectures are supported visit the documentation page. Amazon Bedrock Custom Model Import is generally available in the US-East (N. Virginia), US-West (Oregon), and Europe (Frankfurt) AWS regions.