Qwen 3.6 — Local AI Model by Alibaba Cloud
作者： Jakub Rusinowski · 最后更新： 2026年4月22日
Alibaba's April 2026 follow-up to Qwen 3.5. As of June 15, 2026 only two tiers have been released — a 27B dense model (Apr 21–22) and a 35B-A3B MoE model (Apr 16) — both Apache 2.0, with native text/image/video input and ~256K context (extensible to ~1M via YaRN). Qwen 3.6-Plus/Plus-Preview/Max-Preview exist but are proprietary, API-only, and not listed here.
Licence
| Licence | What it permits | Applies to |
|---|---|---|
Apache-2.0 | Commercial use permitted Commercial use permitted. No usage restrictions beyond attribution. | Qwen 3.6 27B, Qwen 3.6 35B-A3B |
Hardware Requirements
| Qwen 3.6 27B | Min 18 GB VRAM · Q4_K_M · 262,144 ctx · ollama run qwen3.6:27b |
| Qwen 3.6 35B-A3B | Min 22 GB VRAM · Q4_K_M · 262,144 ctx · ollama run qwen3.6:35b-a3b |
Recommended GPU
The cheapest GPU that runs Qwen 3.6 locally (min 18 GB VRAM) is the AMD Radeon RX 7900 XT (20 GB).
How to Run Locally
Install Ollama then run: ollama run qwen3.6:27b
Minimum VRAM: 18 GB. For best results use Q4_K_M quantization.
Qwen 3.6 — Frequently Asked Questions
How much VRAM does Qwen 3.6 need?
Qwen 3.6 needs about 18 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Qwen 3.6 27B (18 GB, Q4_K_M); Qwen 3.6 35B-A3B (22 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run Qwen 3.6 on an RTX 4090 (24 GB)?
Yes — Qwen 3.6 runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
What quantization should I use for Qwen 3.6?
Q4_K_M is the best balance of quality and VRAM for Qwen 3.6 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run Qwen 3.6 with Ollama?
Install Ollama, then run: ollama run qwen3.6:27b. This downloads Qwen 3.6 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.