Alibaba/Qwen · open-weights model
Qwen2.5-32B-Instruct
Qwen2.5-32B-Instruct is a 32.8B-parameter model by Alibaba/Qwen. Its smallest quant (Q4_K_M) needs at least 22.5GB of GPU-addressable memory at 8K context; it loads cleanly at a quality quant on 16 of the 50 machines we track.
- Parameters
- 32.8B
- Layers
- 64
- Max context
- 32K
- Attention
- GQA
- License
- apache-2.0
Config from Qwen/Qwen2.5-32B-Instruct
Quants and what they need
Measured file sizes; “needs” = weights + KV cache at 8K context + overhead.
| Quant | File size | Needs ≥ |
|---|---|---|
| Q8_0 | 34.8 GB | 37.5 GB |
| Q6_K | 26.9 GB | 29.6 GB |
| Q5_K_M | 23.3 GB | 25.9 GB |
| Q4_K_M | 19.9 GB | 22.5 GB |
What runs Qwen2.5-32B-Instruct
Every machine we track, graded at its best clean-fit quant.
Verdicts computed from published specs and measured GGUF sizes — how the math works.