llmrun

Qwen 2.5 Models — Hardware Requirements

41 Qwen 2.5 models from Alibaba and the community, from the smallest that runs in 0.5 GB of VRAM up to 72.7B parameters. Every row links to full quantization tables and GPU compatibility.

All Qwen 2.5 Models by Size

ModelParamsContext
Qwen2.5 0.5B Instruct494M33K
Qwen2.5 Coder 0.5B Instruct494M33K
Qwen2.5 0.5B494M33K
Qwen2.5 Coder 0.5B494M33K
Qwen2.5 1.5B Instruct1.5B33K
Qwen2.5 Coder 1.5B Instruct1.5B33K
Qwen2.5 Math 1.5B Instruct1.5B4K
Qwen2.5 1.5B1.5B131K
Qwen2.5 Math 1.5B1.5B4K
Qwen2.5 Coder 1.5B1.5B33K
Qwen2.5 1.5B Quantized.w8a81.8B33K
Qwen2.5 Omni 3B MNN3B—
Qwen2.5 3B Instruct3.1B33K
Qwen2.5 Coder 3B Instruct3.1B33K
Qwen2.5 3B3.1B33K
Qwen2.5 Coder 3B3.1B33K
Qwen2.5 Coder 3B Claude Opus 4.6 Distilled3.1B33K
Qwen2.5 VL 3B Instruct3.8B128K
Qwen2.5 Omni 3B5.5B—
Ruadapt Qwen2.5 7B Ext U48 Instruct7.6B33K
Qwen2.5 7B Instruct7.6B33K
Qwen2.5 Coder 7B Instruct7.6B33K
Qwen2.5 Math 7B Instruct7.6B4K
Qwen2.5 Coder 7B Instruct Abliterated7.6B33K
Qwen2.5 7B Instruct 1M7.6B1010K
Qwen2.5 7B7.6B131K
Qwen2.5 7B Instruct Uncensored7.6B33K
Qwen2.5 Math 7B7.6B4K
Qwen2.5 Coder 7B7.6B33K
Qwen2.5 Coder 7B Bird Cot7.6B33K
Qwen2.5 VL 7B Instruct8.3B128K
Qwen2.5 Omni 7B10.7B—
Qwen2.5 14B Instruct14.8B33K
Qwen2.5 Coder 14B Instruct14.8B33K
Qwen2.5 14B14.8B131K
Qwen2.5 Coder 14B14.8B33K
Qwen2.5 32B Instruct32.8B33K
Qwen2.5 Coder 32B Instruct32.8B33K
Qwen2.5 Coder 32B32.8B33K
Qwen2.5 32B32.8B131K
Qwen2.5 VL 32B Instruct33.5B128K
Qwen2.5 72B Instruct72.7B33K
Qwen2.5 72B72.7B131K
Qwen2.5 72B Instruct Abliterated72.7B33K

How Qwen 2.5 Compares — Benchmark Rating

Qwen2.5 32B Instruct is the highest-rated Qwen 2.5 model with an overall benchmark rating of 37.2/100 — #53 among 78 open models. The top proprietary model, Claude Fable 5 (high), scores 87.8. Click a model to see its full benchmark breakdown.

Claude Fable 5 (high) · proprietary87.8
Claude Opus 4.7 · proprietary86.4
Claude Fable 5 Max · proprietary82.4
GPT 6 Astra (high) · proprietary82.4
GPT 6 Astra Max · proprietary82.4
Kimi K366.8
Qwen3.5 397B A17B66.1
GLM 565.9
DeepSeek V4.1 Flash65.8
DeepSeek R1 052863.1
Qwen2.5 32B Instruct37.2
Qwen2.5 72B Instruct37.0
Composite of normalized public benchmark scores (methodology) · ■ Qwen 2.5 · ■ other models

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 0.5 GB of VRAM at an aggressive quantization. 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?
40 of 44 Qwen 2.5 models fit in 16 GB of VRAM at some quantization, including Qwen2.5 7B Instruct, Qwen2.5 Coder 7B Instruct, Qwen2.5 14B Instruct.
What is the most popular Qwen 2.5 model to run locally?
Qwen2.5 7B Instruct is the most downloaded Qwen 2.5 model in local-friendly quantized formats. It runs from 2.7 GB of VRAM.
How do Qwen 2.5 models score on benchmarks?
Qwen2.5 32B Instruct leads the family with an overall benchmark rating of 37.2/100, ranking #53 among 78 open models, while the top proprietary model, Claude Fable 5 (high), scores 87.8. See the comparison chart above for the full standings.
Qwen 2.5 Models — VRAM & Hardware Requirements | llmrun