Written by Jakub Rusinowski · Last updated July 10, 2026
Alibaba's model series with a hybrid thinking mode. Qwen 3 models can toggle between fast response mode and deep Chain-of-Thought reasoning on demand — type /think for hard problems, /no_think for fast answers. Available in dense and MoE variants. Previous generation — superseded by Qwen 3.5 and Qwen 3.6, which extend context length and improve reasoning efficiency. Still widely used and supported.
| Qwen 3 8B | Min 6 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3:8b |
| Qwen 3 14B | Min 10 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3:14b |
| Qwen 3 32B | Min 20 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3:32b |
| Qwen 3 30B-A3B (MoE) | Min 8 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3:30b-a3b |
| Qwen 3 235B-A22B (MoE) | Min 80 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3:235b-a22b |
The cheapest GPU that runs Qwen 3 locally (min 6 GB VRAM) is the Intel Arc B570 (10 GB).
Install Ollama then run: ollama run qwen3:8b
Minimum VRAM: 6 GB. For best results use Q4_K_M quantization.
Qwen 3 needs about 6 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Qwen 3 8B (6 GB, Q4_K_M); Qwen 3 14B (10 GB, Q4_K_M); Qwen 3 32B (20 GB, Q4_K_M); Qwen 3 30B-A3B (MoE) (8 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Yes — Qwen 3 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.
Q4_K_M is the best balance of quality and VRAM for Qwen 3 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.
Install Ollama, then run: ollama run qwen3:8b. This downloads Qwen 3 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.