Arc B580 12GB vs Gaudi 3 128GB for Local LLMs

The Gaudi 3 128GB has about 618% more memory bandwidth (3,276 vs 456 GB/s). Batch-1 decode tracks bandwidth closely, so a model that fits both cards should decode roughly 618% faster on the Gaudi 3 128GB. The Gaudi 3 128GB fits more of the catalog at Q4/8K — 47 vs 18 of 61 models (128 vs 12 GB VRAM). Largest model that fits the Gaudi 3 128GB but not the Arc B580 12GB: DeepSeek-V4-Flash (158.1B). MSRP is $249 for the Arc B580 12GB vs $12,500 for the Gaudi 3 128GB. MSRP is a launch list price, not live retail — Amazon prices move; click through for the current price. Not retail-shoppable: Gaudi 3 128GB (datacenter-class; cloud rental is the realistic way to use it). In the Q4/8K head-to-head below (40 shared models), the Gaudi 3 128GB posts the higher tok/s in 40 rows vs 0. Methodology.

Spec comparison

	Arc B580 12GB	Gaudi 3 128GB
Vendor	Intel	Intel
VRAM	12 GB	128 GB
Memory type	GDDR6	HBM2e
Memory bandwidth	456 GB/s	3,276 GB/s
TDP	190 W	600 W
MSRP (list)	$249	$12,500
Buy	Search Amazon (affiliate)	—

MSRP is the launch list price, not live retail. Buy links are Amazon affiliate links — the price after you click is the live price.

Q4 / 8K head-to-head — top 15 of 40 shared models

Model	Params (B)	Arc B580 12GB tok/s	Arc B580 12GB fits	Gaudi 3 128GB tok/s	Gaudi 3 128GB fits	Winner
Qwen3-0.6B	0.8	362.7 est.	Yes	2605.9 est.	Yes	Gaudi 3 128GB
gemma-3-1b-it	1.0	290.2 est.	Yes	2084.7 est.	Yes	Gaudi 3 128GB
Llama-3.2-1B-Instruct	1.2	241.8 est.	Yes	1737.3 est.	Yes	Gaudi 3 128GB
Qwen3-1.7B	2.0	145.1 est.	Yes	1042.4 est.	Yes	Gaudi 3 128GB
Qwen3-30B-A3B	30.5	117.8 est.	No	694.9 est.	Yes	Gaudi 3 128GB
Qwen3-Coder-30B-A3B-Instruct	30.5	117.8 est.	No	694.9 est.	Yes	Gaudi 3 128GB
GLM-4.7-Flash	31.2	113.1 est.	No	694.9 est.	Yes	Gaudi 3 128GB
Qwen3-30B-A3B-Instruct-2507	30.5	117.8 est.	No	694.9 est.	Yes	Gaudi 3 128GB
Qwen3-Next-80B-A3B-Instruct	81.3	18.5 est.	No	694.9 est.	Yes	Gaudi 3 128GB
Qwen3-Coder-Next	79.7	19.3 est.	No	694.9 est.	Yes	Gaudi 3 128GB
Llama-3.2-3B-Instruct	3.2	47.4 measured	Yes	651.5 est.	Yes	Gaudi 3 128GB
gpt-oss-20b	21.5	180.6 est.	No	579.1 est.	Yes	Gaudi 3 128GB
Phi-4-mini-instruct	3.8	76.4 est.	Yes	548.6 est.	Yes	Gaudi 3 128GB
Phi-4-mini-reasoning	3.8	76.4 est.	Yes	548.6 est.	Yes	Gaudi 3 128GB
gemma-3-4b-it	4.3	74.4 est.	Yes	534.5 est.	Yes	Gaudi 3 128GB

est. = bandwidth-based estimate; measured = curated real benchmark that overrides the estimate. Fits = weights + KV(8K) + 1 GB overhead ≤ 95% of VRAM.

Fit coverage at Q4 / 8K

	Arc B580 12GB	Gaudi 3 128GB
Catalog models that fit	18 of 61	47 of 61
Largest exclusive fit	—	DeepSeek-V4-Flash (158.1B)

Arc B580 12GB details → · Gaudi 3 128GB details →