Alibaba/Qwen · open-weights model
Qwen3-30B-A3B
Qwen3-30B-A3B is a 30.5B-parameter model (MoE, 3.3B active) by Alibaba/Qwen. Its smallest quant (Q4_K_M) needs at least 19.9GB of GPU-addressable memory at 8K context; it loads cleanly at a quality quant on 20 of the 50 machines we track.
- Parameters
- 30.5B total · 3.3B active
- Layers
- 48
- Max context
- 40K
- Attention
- GQA
- License
- apache-2.0
Config from Qwen/Qwen3-30B-A3B
Quants and what they need
Measured file sizes; “needs” = weights + KV cache at 8K context + overhead.
| Quant | File size | Needs ≥ |
|---|---|---|
| Q8_0 | 32.5 GB | 33.7 GB |
| Q6_K | 25.1 GB | 26.3 GB |
| Q5_K_M | 21.7 GB | 23.0 GB |
| Q4_K_M | 18.6 GB | 19.9 GB |
What runs Qwen3-30B-A3B
Every machine we track, graded at its best clean-fit quant.
Verdicts computed from published specs and measured GGUF sizes — how the math works.