Qwen3.5-9B
Description
Qwen3.5-9B is a small open language model from Qwen, released on 2026-02-27. It has 9 billion parameters, so it is one of the models that an ordinary computer with a decent graphics card or enough memory can run, especially in a compressed form. It is a previous generation compared with Qwen3.8, but it is much smaller than Qwen3.8-27B.
According to the model card, it understands text, images and video. Thinking mode is on by default and can be switched off to get a direct answer. The context window is 262,144 tokens and can be extended to about one million. The authors say it supports 201 languages and dialects. It works with Transformers, vLLM, SGLang and other tools, and it is widely repackaged for desktop programs that run models locally.
The weights are public without any acceptance step, and the license is Apache-2.0, which allows commercial use. We found no territorial restrictions in the license. The card does not give a video memory requirement. In the authors' own table the model is compared with larger open models; these are the authors' figures, not independent tests.
Date of check: 2026-10-21.
Pros and cons
Pros
- Small: 9 billion parameters, within reach of an ordinary computer in a compressed form
- Apache-2.0 license, weights open without an acceptance step
- Understands text, images and video
- Thinking mode that can be switched off
- Context of 262,144 tokens, 201 languages and dialects per the authors
Cons
- Previous generation: Qwen3.8 is newer
- Weaker than larger open models
- No audio input
- The card does not state hardware requirements
- Benchmark figures come from the authors, not from independent tests
Key facts
- License
- Apache-2.0
- Commercial use
- Allowed
- Developer
- Qwen
- Interface languages
- English
- Checked on
- Oct 21, 2026