Qwen Releases 9B Multimodal Model in New 3.5 Series

The Qwen team at Alibaba has introduced Qwen3.5-9B, the first release in its new generation of open-source models. This 9-billion parameter model is primarily a vision-language model (VLM), designed to understand and process both text and images to generate text-based outputs.

As a dense model, Qwen3.5-9B offers a powerful and efficient architecture in a popular size class. It operates with a context window of 8,192 tokens, making it suitable for a range of multimodal tasks. Importantly, the model is released under the permissive Apache 2.0 license, allowing for broad commercial and research applications without restrictive usage policies.

The release provides developers with a capable, mid-sized multimodal model that balances performance with computational efficiency. Its open license and manageable size make it an attractive option for teams building applications that require visual understanding, such as advanced chatbots, content generation tools, and accessibility software, without needing the resources to run massive, proprietary models.

Developers can explore the model's capabilities and access the weights directly from its Hugging Face repository. The release continues the Qwen team's consistent contribution of high-quality, open models to the AI community.