Can GB200 NVL72 run Vicuna 33B?

Yes, the GB200 NVL72 has 13824GB VRAM and Vicuna 33B requires approximately 18.2GB VRAM for Q4 quantization. Estimated inference speed: 152 tokens/second.

✅ — GB200 NVL72 Can Run Vicuna 33B

The GB200 NVL72 packs 13824GB VRAM. Vicuna 33B needs ~18.2GB. 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: Vicuna 33B
Parameters: 33B
VRAM (Q4): ~18.2 GB
Context: 2,048

Estimated Performance

152 tok/s

Estimated inference speed (Q4 quantization)

Run it locally with Ollama

ollama run vicuna:33b

Requires Ollama installed on your machine

View on Ollama

Vicuna 33B Details GB200 NVL72 Specs

Try Vicuna 33B on a different GPU

Try another model on GB200 NVL72

Full compatibility matrix

✅ — GB200 NVL72 Can Run Vicuna 33B

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try Vicuna 33B 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 Vicuna 33B

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try Vicuna 33B on a different GPU

Try another model on GB200 NVL72

Related Checks

More tools

VRAM Calculator

Cloud GPU Pricing

GPU Comparison

Compatibility Lab