Qwen Models — Hardware Requirements
28 Qwen models from Alibaba and the community, from the smallest that runs in 1.6 GB of VRAM at Q4 up to 111.2B parameters. Every row links to full quantization tables and GPU compatibility.
All Qwen Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Qwen1.5 0.5B Chat | 620M | 2.3 GB | 33K | ||
| Qwen1.5 0.5B | 620M | 2.3 GB | 33K | ||
| Qwen Base Invoicev1.01 1.5B | 1.5B | 1.7 GB | 33K | ||
| Qwen1.5 1.8B Chat | 1.8B | 4.6 GB | 33K | ||
| Qwen 1 8B | 1.8B | 1.9 GB | 8K | ||
| Qwen1.5 1.8B | 1.8B | 4.6 GB | 33K | ||
| Qwen1.5 4B | 4.0B | 9.4 GB | 33K | ||
| CodeQwen1.5 7B | 7.3B | 5.7 GB | 66K | ||
| Qwen 7B Chat | 7.7B | 7.6 GB | 33K | ||
| Qwen1.5 7B | 7.7B | 13.5 GB | 33K | ||
| Qwen1.5 7B Chat | 7.7B | 13.5 GB | 33K | ||
| Qwen 7B | 7.7B | 7.6 GB | 33K | ||
| Qwen Image 2.1 PE T2I | 9.4B | 6.5 GB | 262K | ||
| Qwen 14B Chat | 14.2B | 10.6 GB | 8K | ||
| Qwen1.5 14B Chat | 14.2B | 22.2 GB | 33K | ||
| Qwen 14B | 14.2B | 10.6 GB | 8K | ||
| Qwen1.5 14B | 14.2B | 22.2 GB | 33K | ||
| Qwen1.5 MoE A2.7B | 14.3B | 10.5 GB | 8K | ||
| Qwen1.5 MoE A2.7B Chat | 14.3B | 12.1 GB | 33K | ||
| Qwen1.5 32B | 32.5B | 24.1 GB | 33K | ||
| Qwen1.5 32B Chat | 32.5B | 24.1 GB | 33K | ||
| Qwen AgentWorld 35B A3B | 34.7B | 21.3 GB | 262K | ||
| Qwen 72B | 72.3B | 52.5 GB | 33K | ||
| Qwen1.5 72B Chat | 72.3B | 86.6 GB | 33K | ||
| Qwen1.5 110B Chat | 111.2B | 72.4 GB | 33K | ||
| Qwen1.5 110B | 111.2B | 72.4 GB | 33K |
Community variants and fine-tunes(11)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| SpatialLM1.1 Qwen 0.5B | 604M | 1.6 GB | 33K | ||
| Nemotron Research Reasoning Qwen 1.5B | 1.8B | 1.8 GB | 131K | ||
| Qwen35 4B Soyuz Merged | 4B | 4.8 GB | 262K | ||
| CyberSecQwen 4B | 4.0B | 4.2 GB | 262K | ||
| ELYZA Shortcut 1.0 Qwen 7B | 7.6B | 5.8 GB | 33K | ||
| Qwen Marketing | 8.2B | 8.0 GB | — | ||
| Qwen27b Abliterated Fable MTP | 27B | 21.4 GB | — | ||
| Cogito V1 Preview Qwen 32B | 32B | 23.8 GB | 131K | ||
| XiYanSQL QwenCoder 32B 2504 | 32B | 23.8 GB | 33K | ||
| Qwen35B Agent R2 | 34.7B | 21.3 GB | 262K | ||
| Qwen35b Agent R2O3 | 34.7B | 21.3 GB | 262K |
Frequently Asked Questions
- How much VRAM do I need to run a Qwen model?
- The smallest Qwen model, SpatialLM1.1 Qwen 0.5B, runs from 1.6 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 1.5 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which Qwen models can I run on a 16 GB GPU?
- 23 of 37 Qwen models fit in 16 GB of VRAM at Q4_K_M, including Qwen Image 2.1 PE T2I, Qwen1.5 0.5B Chat, Qwen 7B Chat.
- What is the most popular Qwen model to run locally?
- Qwen AgentWorld 35B A3B is the most downloaded Qwen model in local-friendly quantized formats. It runs from 21.3 GB of VRAM at Q4_K_M.