Qwen 3.5 9B
by Alibaba · qwen-3.5 family
9B
parameters
text-generation code-generation reasoning multilingual vision tool-use math
Qwen 3.5 9B is the sweet spot of the small Qwen 3.5 lineup — strong at coding, math, reasoning, and vision tasks while staying under 8 GB VRAM at Q4. It supports dual thinking modes for flexible speed/quality tradeoffs. A great all-rounder for 12-16 GB GPUs or 16 GB Macs where you want multimodal capability without heavy resource requirements.
Quick Start with Ollama
ollama run 9b-q4_K_M | Creator | Alibaba |
| Parameters | 9B |
| Architecture | transformer-decoder |
| Context | 256K tokens |
| Released | Mar 2, 2026 |
| License | Apache 2.0 |
| Ollama | qwen3.5:9b |
Quantization Options
| Format | File Size | VRAM Required | Quality | Ollama Tag |
|---|---|---|---|---|
| Q4_K_M rec | 6.6 GB | 7.5 GB | | 9b-q4_K_M |
| Q8_0 | 10 GB | 11.5 GB | | 9b-q8_0 |
Compatible Hardware
Q4_K_M requires 7.5 GB VRAM