Can GB200 NVL72 run Gemma 2 27B?

Yes, the GB200 NVL72 has 13824GB VRAM and Gemma 2 27B requires approximately 14.9GB VRAM for Q4 quantization. Estimated inference speed: 186 tokens/second.

✅ — GB200 NVL72 Can Run Gemma 2 27B

The GB200 NVL72 packs 13824GB VRAM. Gemma 2 27B needs ~14.9GB. 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: Gemma 2 27B
Parameters: 27B
VRAM (Q4): ~14.9 GB
Context: 8,192

Estimated Performance

186 tok/s

Estimated inference speed (Q4 quantization)

Run it locally with Ollama

ollama run gemma2:27b

Requires Ollama installed on your machine

View on Ollama

Gemma 2 27B Details GB200 NVL72 Specs

Try Gemma 2 27B on a different GPU

Try another model on GB200 NVL72

Full compatibility matrix

✅ — GB200 NVL72 Can Run Gemma 2 27B

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try Gemma 2 27B 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 Gemma 2 27B

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try Gemma 2 27B on a different GPU

Try another model on GB200 NVL72

Related Checks

More tools

VRAM Calculator

Cloud GPU Pricing

GPU Comparison

Compatibility Lab