4 new Qwen models for multimodal reasoning, agentic coding, and multilingual applications are now available in Amazon SageMaker JumpStart

This article announces the availability of four new Qwen models in Amazon SageMaker JumpStart for various AI applications.

• Qwen3.5-27B-FP8: Multimodal vision-language model supporting 1M token context and 200+ languages
• Qwen3.6-35B-A3B: Efficient Mixture-of-Experts model optimized for agentic coding workflows
• Qwen3.5-0.8B: Compact multimodal model for edge deployments and on-device inference
• Qwen3.5-2B: Lightweight multimodal model for prototyping and moderate-compute tasks
• All models deployable via SageMaker Studio or Python SDK with minimal setup

These models enable developers to build multimodal reasoning, agentic coding, and multilingual applications with varying compute requirements.