Qwen 2.5 Models — Hardware Requirements
44 Qwen 2.5 models from Alibaba and the community, from the smallest that runs in 0.8 GB of VRAM at Q4 up to 73.4B parameters. Every row links to full quantization tables and GPU compatibility.
All Qwen 2.5 Models by Size
Community variants and fine-tunes(7)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| Qwen2.5 1.5B Quantized.w8a8 | 1.8B | 1.8 GB | 33K | ||
| Qwen2.5 Omni 3B MNN | 3B | 3.9 GB | — | ||
| Qwen2.5 Coder 3B Claude Opus 4.6 Distilled | 3.1B | 2.8 GB | 33K | ||
| Qwen2.5 Coder 7B Instruct Abliterated | 7.6B | 5.8 GB | 33K | ||
| Qwen2.5 7B Instruct Uncensored | 7.6B | 5.8 GB | 33K | ||
| Qwen2.5 Coder 7B Bird Cot | 7.6B | 5.8 GB | 33K | ||
| Qwen2.5 72B Instruct Abliterated | 72.7B | 49.3 GB | 33K |
How Qwen 2.5 Compares — llmrun Composite Score
Qwen2.5 72B Instruct is the highest-scoring Qwen 2.5 model with an llmrun Composite Score of 82.2 — #84 among 106 open models. The top proprietary model, Gemini 4 Carbon, scores 158.6. Click a model to see its full benchmark breakdown.
Gemini 4 Carbon · proprietary158.6
GPT 6 Astra · proprietary156.2
Claude Opus 5.5 · proprietary153.0
GPT 6.1 Sol · proprietary151.7
GPT 6.1 Terra · proprietary150.1
Kimi K3.1 Pro142.8
MiMo V2.6 Pro RL133.3
Kimi K3132.7
MiMo V2.6 Flash RL131.9
GLM 5.3130.1
DeepSeek V4.1 Flash129.4
DeepSeek V4 Pro 0813129.2
Frequently Asked Questions
- How much VRAM do I need to run a Qwen 2.5 model?
- The smallest Qwen 2.5 model, Qwen2.5 0.5B Instruct, runs from 0.8 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 0.5 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which Qwen 2.5 models can I run on a 16 GB GPU?
- 38 of 47 Qwen 2.5 models fit in 16 GB of VRAM at Q4_K_M, including Qwen2.5 VL 7B Instruct, Qwen2.5 7B Instruct, Qwen2.5 Coder 7B Instruct.
- What is the most popular Qwen 2.5 model to run locally?
- Qwen2.5 VL 7B Instruct is the most downloaded Qwen 2.5 model in local-friendly quantized formats. It runs from 6.2 GB of VRAM at Q4_K_M.
- How do Qwen 2.5 models score on benchmarks?
- Qwen2.5 72B Instruct leads the family with an llmrun Composite Score of 82.2, ranking #84 among 106 open models, while the top proprietary model, Gemini 4 Carbon, scores 158.6. See the comparison chart above for the full standings.