What GPU do you need to run Nemotron 3.5 Lightning 30B A3B?
31.58B params 256K max context Tool useReasoning
The smallest GPU memory that runs Nemotron 3.5 Lightning 30B A3B well is 24 GB (Q4_K_M, e.g. RTX Spark). 45 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 81.94 SWE-bench Verified 51.56
VRAM needed per quantization
| Quantization | Weights | 16K context | 32K context | Full 256K |
|---|---|---|---|---|
| Q4_K_M | 23.7 GB | 23.8 GB | 23.9 GB | 25.3 GB |
| Q8_0 | 32.6 GB | 32.7 GB | 32.8 GB | 34.1 GB |
Weights plus fp16 KV cache. Leave ~1 GB headroom for the inference engine.
Cheapest GPUs that can run Nemotron 3.5 Lightning 30B 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 Nemotron 3.5 Lightning 30B 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 3090 24 GB Q4_K_M 44K context · ~288 tok/s
- RTX 3090 Ti 24 GB Q4_K_M 44K context · ~310 tok/s
- RTX 4090 24 GB Q4_K_M 44K context · ~310 tok/s
- RTX 4500 Ada 24 GB Q4_K_M 44K context · ~133 tok/s
- RTX A5000 24 GB Q4_K_M 44K context · ~236 tok/s
- RTX PRO 4000 Blackwell 24 GB Q4_K_M 44K context · ~207 tok/s
- RTX Spark 24 GB Q4_K_M 44K context · ~92 tok/s
- Tesla P40 24 GB Q4_K_M 44K context · ~106 tok/s
- RTX 5090 32 GB Q4_K_M 256K context · ~551 tok/s
- RTX 5000 Ada 32 GB Q4_K_M 256K context · ~177 tok/s
- RTX PRO 4500 Blackwell 32 GB Q4_K_M 256K context · ~276 tok/s
- Tesla V100 32GB 32 GB Q4_K_M 256K context · ~277 tok/s
- L40S 48 GB Q8_0 256K context · ~211 tok/s
- RTX 6000 Ada 48 GB Q8_0 256K context · ~234 tok/s
- RTX A6000 48 GB Q8_0 256K context · ~187 tok/s
- RTX PRO 5000 Blackwell 48 GB Q8_0 256K context · ~328 tok/s
- RTX PRO 5000 Blackwell 72GB 72 GB Q8_0 256K context · ~328 tok/s
- RTX PRO 5500 Blackwell 84 GB Q8_0 256K context · ~341 tok/s
- RTX PRO 6000 Blackwell 96 GB Q8_0 256K context · ~437 tok/s
- DGX Spark (GB 10) 128 GB Q8_0 256K context · ~67 tok/s
AMD
- Radeon RX 7900 XTX 24 GB Q4_K_M 44K context · ~295 tok/s
- Radeon AI PRO R9700 32 GB Q4_K_M 256K context · ~197 tok/s
- Ryzen AI Max (Strix Halo) 32 GB Q4_K_M 256K context · ~79 tok/s
- Ryzen AI Max 400 (Gorgon Halo) 64 GB Q8_0 256K context · ~67 tok/s
Intel
- Arc Pro B65 32 GB Q4_K_M 256K context · ~187 tok/s
- Arc Pro B70 32 GB Q4_K_M 256K context · ~187 tok/s
Apple
- M2 24 GB Q4_K_M 44K context · ~31 tok/s
- M3 24 GB Q4_K_M 44K context · ~31 tok/s
- M4 24 GB Q4_K_M 44K context · ~37 tok/s
- M4 Pro 24 GB Q4_K_M 44K context · ~84 tok/s
- M5 24 GB Q4_K_M 44K context · ~47 tok/s
- M5 Pro 24 GB Q4_K_M 44K context · ~94 tok/s
- M6 24 GB Q4_K_M 44K context · ~52 tok/s
- M1 Max 32 GB Q4_K_M 256K context · ~123 tok/s
- M1 Pro 32 GB Q4_K_M 256K context · ~62 tok/s
- M2 Max 32 GB Q4_K_M 256K context · ~123 tok/s
- M2 Pro 32 GB Q4_K_M 256K context · ~62 tok/s
- M3 Max 36 GB Q8_0 256K context · ~73 tok/s
- M3 Pro 36 GB Q8_0 256K context · ~37 tok/s
- M4 Max 36 GB Q8_0 256K context · ~100 tok/s
- M5 Max 36 GB Q8_0 256K context · ~112 tok/s
- M1 Ultra 64 GB Q8_0 256K context · ~195 tok/s
- M2 Ultra 64 GB Q8_0 256K context · ~195 tok/s
- M3 Ultra 96 GB Q8_0 256K context · ~200 tok/s
- M5 Ultra 96 GB Q8_0 256K context · ~293 tok/s