Alibaba has introduced the new open source Qwen3.5 series built for native multimodal agents. The first model in this series is a ~400B parameter native vision…

NVIDIA DevTalk serves as a vibrant community hub where developers can engage in discussions, seek assistance, and collaborate on projects involving NVIDIA hardware and software. Developers can tap into the collective expertise of the NVIDIA developer community, sharing insights, troubleshooting issues, and exploring best practices for GPU programming and AI development. Additionally, DevTalk provides a platform for developers to showcase their projects, receive feedback, and network with peers, fostering collaboration and knowledge exchange within the NVIDIA ecosystem.

NVIDIA Developer

Alibaba's Qwen3.5 is a ~400B parameter open-source vision-language model (VLM) built for native multimodal agents, using a hybrid MoE and Gated Delta Networks architecture with 17B active parameters and a 256K context window extensible to 1M tokens. It supports UI navigation, visual reasoning, coding, and complex search across 200+ languages. NVIDIA provides GPU-accelerated endpoints via build.nvidia.com (powered by Blackwell GPUs) with free API access through the NVIDIA Developer Program. NVIDIA NIM enables containerized production deployment, while the NeMo Automodel library supports fine-tuning via SFT or LoRA, including a Medical Visual QA reference tutorial on radiological datasets. Multinode Slurm and Kubernetes deployments are supported for large-scale MoE workloads.

Develop Native Multimodal Agents with Qwen3.5 VLM Using NVIDIA GPU-Accelerated Endpoints