can-i-run-this-llm

GPU comparison

NVIDIA RTX Pro 6000 Blackwell (96GB) vs Intel Arc B580 (12GB) for local LLMs

How the two stack up for running open-source LLMs locally — memory, bandwidth, price, and how many of the 34 tracked models each one can run.

NVIDIA RTX Pro 6000 Blackwell (96GB)Intel Arc B580 (12GB)
Memory96.0 GB12.0 GB
Bandwidth1792 GB/s456 GB/s
Price (approx)$12,999$289
LLMs it runs30 of 3413 of 34
Best model it runsGemma 4 31B · 20–30 tok/sDevstral Small 2 24B · 32–48 tok/s

The NVIDIA RTX Pro 6000 Blackwell (96GB) runs 17 more of the tracked models (30 vs 13), thanks to its 96.0 GB of memory. The NVIDIA RTX Pro 6000 Blackwell (96GB) has more memory bandwidth (1792 GB/s), so it generates tokens faster at the same model and quant.

What the NVIDIA RTX Pro 6000 Blackwell (96GB) runs →What the Intel Arc B580 (12GB) runs →Open the calculator →