Can GB200 NVL72 run Granite 8B Code?

Yes, the GB200 NVL72 has 13824GB VRAM and Granite 8B Code requires approximately 4.4GB VRAM for Q4 quantization. Estimated inference speed: 628 tokens/second.

✅ — GB200 NVL72 Can Run Granite 8B Code

The GB200 NVL72 packs 13824GB VRAM. Granite 8B Code needs ~4.4GB. 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: Granite 8B Code
Parameters: 8B
VRAM (Q4): ~4.4 GB
Context: 8,192

Estimated Performance

628 tok/s

Estimated inference speed (Q4 quantization)

Run it locally with Ollama

ollama run granite-code:8b

Requires Ollama installed on your machine

View on Ollama

Granite 8B Code Details GB200 NVL72 Specs

Try Granite 8B Code on a different GPU

Try another model on GB200 NVL72

Full compatibility matrix

✅ — GB200 NVL72 Can Run Granite 8B Code

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try Granite 8B Code 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 Granite 8B Code

Hardware Specs

Model Requirements

Estimated Performance

Run it locally with Ollama

Try Granite 8B Code on a different GPU

Try another model on GB200 NVL72

Related Checks

More tools

VRAM Calculator

Cloud GPU Pricing

GPU Comparison

Compatibility Lab