What GPU do you need to run Qwen3 Next 80B-A3B?
80B params 256K max context Tool use
The smallest GPU memory that runs Qwen3 Next 80B-A3B well is 64 GB (Q4_K_M, e.g. RTX Spark). 18 of 132 GPUs run it well. It's a mixture-of-experts model, so all weights must fit in memory but speed depends only on the active experts.
MMLU 90
VRAM needed per quantization
| Quantization | Weights | 16K context | 32K context | Full 256K |
|---|---|---|---|---|
| Q4_K_M | 48.4 GB | 48.8 GB | 49.1 GB | 54.3 GB |
| Q8_0 | 84.8 GB | 85.2 GB | 85.5 GB | 90.7 GB |
Weights plus fp16 KV cache. Leave ~1 GB headroom for the inference engine.
Cheapest GPUs that can run Qwen3 Next 80B-A3B
Lowest launch MSRP among GPUs that run its smallest quantization (Q4_K_M) well, with at least 16K context.
As an Amazon Associate, I earn from qualifying purchases.
GPUs that can run Qwen3 Next 80B-A3B
Best quantization that fits on each GPU, with the context and speed it gets. For chips with several memory sizes, the smallest that works is shown.
NVIDIA
- RTX Spark 64 GB Q4_K_M 256K context · ~107 tok/s
- RTX PRO 5000 Blackwell 72GB 72 GB Q4_K_M 256K context · ~478 tok/s
- RTX PRO 5500 Blackwell 84 GB Q4_K_M 256K context · ~498 tok/s
- RTX PRO 6000 Blackwell 96 GB Q8_0 256K context · ~429 tok/s
- DGX Spark (GB 10) 128 GB Q8_0 256K context · ~65 tok/s
AMD
- Ryzen AI Max (Strix Halo) 64 GB Q4_K_M 256K context · ~91 tok/s
- Ryzen AI Max 400 (Gorgon Halo) 64 GB Q4_K_M 256K context · ~97 tok/s
Apple
- M1 Max 64 GB Q4_K_M 256K context · ~142 tok/s
- M1 Ultra 64 GB Q4_K_M 256K context · ~285 tok/s
- M2 Max 64 GB Q4_K_M 256K context · ~142 tok/s
- M2 Ultra 64 GB Q4_K_M 256K context · ~285 tok/s
- M3 Max 64 GB Q4_K_M 256K context · ~142 tok/s
- M4 Max 64 GB Q4_K_M 256K context · ~194 tok/s
- M4 Pro 64 GB Q4_K_M 256K context · ~97 tok/s
- M5 Max 64 GB Q4_K_M 256K context · ~219 tok/s
- M5 Pro 64 GB Q4_K_M 256K context · ~109 tok/s
- M3 Ultra 96 GB Q8_0 256K context · ~196 tok/s
- M5 Ultra 96 GB Q8_0 256K context · ~287 tok/s