Can GB200 NVL72 run SmolLM 1.7B?

Yes, the GB200 NVL72 has 13824GB VRAM and SmolLM 1.7B requires approximately 0.9GB VRAM for Q4 quantization. Estimated inference speed: 2956 tokens/second.

✅ — GB200 NVL72 Can Run SmolLM 1.7B

The GB200 NVL72 packs 13824GB VRAM. SmolLM 1.7B needs ~0.9GB. You're good to go.

✅ Plenty of headroom

Hardware Specs

GPU: GB200 NVL72
VRAM: 13824 GB
Bandwidth: 3350 GB/s
MSRP: $0

Model Requirements

Model: SmolLM 1.7B
Parameters: 1.7B
VRAM (Q4): ~0.9 GB
Context: 8,192

Estimated Performance

2956 tok/s

Estimated inference speed (Q4 quantization)

Run it locally with Ollama

ollama run smollm:1.7b

Requires Ollama installed on your machine

View on Ollama

SmolLM 1.7B Details GB200 NVL72 Specs

Try SmolLM 1.7B on a different GPU

Try another model on GB200 NVL72

Full compatibility matrix

✅ — GB200 NVL72 Can Run SmolLM 1.7B

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try SmolLM 1.7B on a different GPU

Try another model on GB200 NVL72

Related Checks

More tools

VRAM Calculator

Cloud GPU Pricing

GPU Comparison

Compatibility Lab

✅ — GB200 NVL72 Can Run SmolLM 1.7B

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try SmolLM 1.7B on a different GPU

Try another model on GB200 NVL72

Related Checks

More tools

VRAM Calculator

Cloud GPU Pricing

GPU Comparison

Compatibility Lab