RTX 4080 Super 16GB vs RTX 4090D 24GB (China) for Local LLMs

The RTX 4090D 24GB (China) has about 37% more memory bandwidth (1,008 vs 736 GB/s). Batch-1 decode tracks bandwidth closely, so a model that fits both cards should decode roughly 37% faster on the RTX 4090D 24GB (China). The RTX 4090D 24GB (China) fits more of the catalog at Q4/8K — 33 vs 25 of 61 models (24 vs 16 GB VRAM). Largest model that fits the RTX 4090D 24GB (China) but not the RTX 4080 Super 16GB: GLM-4.7-Flash (31.2B). MSRP is $999 for the RTX 4080 Super 16GB vs $1,599 for the RTX 4090D 24GB (China). MSRP is a launch list price, not live retail — Amazon prices move; click through for the current price. Not retail-shoppable: RTX 4090D 24GB (China) (datacenter-class; cloud rental is the realistic way to use it). In the Q4/8K head-to-head below (44 shared models), the RTX 4090D 24GB (China) posts the higher tok/s in 35 rows vs 9. Methodology.

Spec comparison

	RTX 4080 Super 16GB	RTX 4090D 24GB (China)
Vendor	NVIDIA	NVIDIA
VRAM	16 GB	24 GB
Memory type	GDDR6X	GDDR6X
Memory bandwidth	736 GB/s	1,008 GB/s
TDP	320 W	425 W
MSRP (list)	$999	$1,599
Buy	ASUS ProArt GeForce RTX 4080 Super OC (affiliate) · More listings	—

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 44 shared models

Model	Params (B)	RTX 4080 Super 16GB tok/s	RTX 4080 Super 16GB fits	RTX 4090D 24GB (China) tok/s	RTX 4090D 24GB (China) fits	Winner
Qwen3-0.6B	0.8	585.5 est.	Yes	801.8 est.	Yes	RTX 4090D 24GB (China)
gemma-3-1b-it	1.0	468.4 est.	Yes	641.5 est.	Yes	RTX 4090D 24GB (China)
Llama-3.2-1B-Instruct	1.2	390.3 est.	Yes	534.5 est.	Yes	RTX 4090D 24GB (China)
Qwen3-30B-A3B	30.5	338.0 est.	No	213.8 est.	Yes	RTX 4080 Super 16GB
Qwen3-Coder-30B-A3B-Instruct	30.5	338.0 est.	No	213.8 est.	Yes	RTX 4080 Super 16GB
Qwen3-30B-A3B-Instruct-2507	30.5	338.0 est.	No	213.8 est.	Yes	RTX 4080 Super 16GB
GLM-4.7-Flash	31.2	324.5 est.	No	213.8 est.	Yes	RTX 4080 Super 16GB
Qwen3-1.7B	2.0	234.2 est.	Yes	320.7 est.	Yes	RTX 4090D 24GB (China)
Llama-3.2-3B-Instruct	3.2	146.4 est.	Yes	200.5 est.	Yes	RTX 4090D 24GB (China)
gpt-oss-20b	21.5	186.5 measured	Yes	178.2 est.	Yes	RTX 4080 Super 16GB
Qwen3-Coder-Next	79.7	55.3 est.	No	170.4 est.	No	RTX 4090D 24GB (China)
Phi-4-mini-instruct	3.8	123.3 est.	Yes	168.8 est.	Yes	RTX 4090D 24GB (China)
Phi-4-mini-reasoning	3.8	123.3 est.	Yes	168.8 est.	Yes	RTX 4090D 24GB (China)
gemma-3-4b-it	4.3	120.1 est.	Yes	164.5 est.	Yes	RTX 4090D 24GB (China)
Qwen3-Next-80B-A3B-Instruct	81.3	53.2 est.	No	164.0 est.	No	RTX 4090D 24GB (China)

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 4080 Super 16GB	RTX 4090D 24GB (China)
Catalog models that fit	25 of 61	33 of 61
Largest exclusive fit	—	GLM-4.7-Flash (31.2B)

RTX 4080 Super 16GB details → · RTX 4090D 24GB (China) details →