RTX A6000 48GB vs Mac Studio M3 Ultra 96GB for Local LLMs

The Mac Studio M3 Ultra 96GB has about 7% more memory bandwidth (819 vs 768 GB/s). Batch-1 decode tracks bandwidth closely, so a model that fits both cards should decode roughly 7% faster on the Mac Studio M3 Ultra 96GB. The Mac Studio M3 Ultra 96GB fits more of the catalog at Q4/8K — 47 vs 39 of 61 models (96 vs 48 GB VRAM). Largest model that fits the Mac Studio M3 Ultra 96GB but not the RTX A6000 48GB: DeepSeek-V4-Flash (158.1B). MSRP is $4,500 for the RTX A6000 48GB vs $3,999 for the Mac Studio M3 Ultra 96GB. MSRP is a launch list price, not live retail — Amazon prices move; click through for the current price. In the Q4/8K head-to-head below (50 shared models), the Mac Studio M3 Ultra 96GB posts the higher tok/s in 42 rows vs 8. Methodology.

Spec comparison

	RTX A6000 48GB	Mac Studio M3 Ultra 96GB
Vendor	NVIDIA	Apple
VRAM	48 GB	96 GB
Memory type	GDDR6	LPDDR5
Memory bandwidth	768 GB/s	819 GB/s
TDP	300 W	480 W
MSRP (list)	$4,500	$3,999
Buy	Search Amazon (affiliate)	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 50 shared models

Model	Params (B)	RTX A6000 48GB tok/s	RTX A6000 48GB fits	Mac Studio M3 Ultra 96GB tok/s	Mac Studio M3 Ultra 96GB fits	Winner
Qwen3-0.6B	0.8	610.9 est.	Yes	651.5 est.	Yes	Mac Studio M3 Ultra 96GB
gemma-3-1b-it	1.0	488.7 est.	Yes	521.2 est.	Yes	Mac Studio M3 Ultra 96GB
Qwen3-Next-80B-A3B-Instruct	81.3	499.8 est.	No	173.7 est.	Yes	RTX A6000 48GB
Llama-3.2-1B-Instruct	1.2	407.3 est.	Yes	434.3 est.	Yes	Mac Studio M3 Ultra 96GB
Qwen3-1.7B	2.0	244.4 est.	Yes	260.6 est.	Yes	Mac Studio M3 Ultra 96GB
Qwen3-30B-A3B	30.5	162.9 est.	Yes	173.7 est.	Yes	Mac Studio M3 Ultra 96GB
Qwen3-Coder-30B-A3B-Instruct	30.5	162.9 est.	Yes	173.7 est.	Yes	Mac Studio M3 Ultra 96GB
GLM-4.7-Flash	31.2	162.9 est.	Yes	173.7 est.	Yes	Mac Studio M3 Ultra 96GB
Qwen3-30B-A3B-Instruct-2507	30.5	162.9 est.	Yes	173.7 est.	Yes	Mac Studio M3 Ultra 96GB
Qwen3-Coder-Next	79.7	162.9 est.	Yes	173.7 est.	Yes	Mac Studio M3 Ultra 96GB
Llama-3.2-3B-Instruct	3.2	152.7 est.	Yes	154.6 measured	Yes	Mac Studio M3 Ultra 96GB
Phi-4-mini-instruct	3.8	128.6 est.	Yes	137.2 est.	Yes	Mac Studio M3 Ultra 96GB
Phi-4-mini-reasoning	3.8	128.6 est.	Yes	137.2 est.	Yes	Mac Studio M3 Ultra 96GB
gpt-oss-20b	21.5	135.8 est.	Yes	115.5 measured	Yes	RTX A6000 48GB
gpt-oss-120b	120.4	135.5 est.	No	102.2 est.	Yes	RTX A6000 48GB

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

	RTX A6000 48GB	Mac Studio M3 Ultra 96GB
Catalog models that fit	39 of 61	47 of 61
Largest exclusive fit	—	DeepSeek-V4-Flash (158.1B)

RTX A6000 48GB details → · Mac Studio M3 Ultra 96GB details →