Vision
Qwen 3.5 Small Models: The 9B Beats Last-Gen 30B — Here's What Matters for Local AI
Alibaba's Qwen 3.5 drops 4 small models (0.8B to 9B) — all natively multimodal, 262K context, Apache 2.0. The 9B beats Qwen3-30B on reasoning and destroys GPT-5-Nano on vision. VRAM tables and what to run.
Qwen 3.5 9B: Setup, Real-World Testing, and Why It's the New Default for 8GB GPUs
Get Qwen 3.5 9B running in 2 minutes flat. Quantization table, VRAM by GPU tier, thinking mode toggle, native vision testing, and honest head-to-heads vs Llama 3.2, Gemma 3, and Phi-4-mini.