B200 192GB SXM vs V100 16GB PCIe for Local LLMs

The B200 192GB SXM has about 789% more memory bandwidth (8,000 vs 900 GB/s). Batch-1 decode tracks bandwidth closely, so a model that fits both cards should decode roughly 789% faster on the B200 192GB SXM. The B200 192GB SXM fits more of the catalog at Q4/8K — 61 vs 25 of 73 models (192 vs 16 GB VRAM). Largest model that fits the B200 192GB SXM but not the V100 16GB PCIe: MiMo-V2.5 (310.0B). MSRP is $40,000 for the B200 192GB SXM vs $8,000 for the V100 16GB PCIe. MSRP is a launch list price, not live retail — Amazon prices move; click through for the current price. Not retail-shoppable: B200 192GB SXM, V100 16GB PCIe (datacenter-class; cloud rental is the realistic way to use it). In the Q4/8K head-to-head below (49 shared models), the B200 192GB SXM posts the higher tok/s in 49 rows vs 0. Methodology.

Spec comparison

	B200 192GB SXM	V100 16GB PCIe
Vendor	NVIDIA	NVIDIA
VRAM	192 GB	16 GB
Memory type	HBM3e	HBM2
Memory bandwidth	8,000 GB/s	900 GB/s
TDP	1000 W	250 W
MSRP (list)	$40,000	$8,000
Buy	—	—

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 49 shared models

Model	Params (B)	B200 192GB SXM tok/s	B200 192GB SXM fits	V100 16GB PCIe tok/s	V100 16GB PCIe fits	Winner
MiniCPM5-1B	1.08	5000 est.	Yes	895.6 est.	Yes	B200 192GB SXM
Qwen3.5-2B	2.0	4341.1 est.	Yes	483.6 est.	Yes	B200 192GB SXM
DeepSeek-R1-0528-Qwen3-8B-layer-mix-bpw-3.8-mlx	2.3	3774.9 est.	Yes	420.5 est.	Yes	B200 192GB SXM
Qwen3.6-35B-A3B	36.0	2894.1 est.	Yes	316.9 est.	No	B200 192GB SXM
Qwen3.5-35B-A3B	34.7	2894.1 est.	Yes	339.8 est.	No	B200 192GB SXM
GLM-4.7-Flash	31.2	2894.1 est.	Yes	313.0 est.	No	B200 192GB SXM
Qwen3-30B-A3B-Instruct-2507	30.5	2894.1 est.	Yes	405.4 est.	No	B200 192GB SXM
Nemotron-3-Nano-Omni-30B-A3B-Reasoning-BF16	30.0	2894.1 est.	Yes	434.0 est.	No	B200 192GB SXM
Qwen3-Coder-Next	79.7	2894.1 est.	Yes	67.0 est.	No	B200 192GB SXM
Ornith-1.0-35B	35.0	2894.1 est.	Yes	334.2 est.	No	B200 192GB SXM
Qwen3-Next-80B-A3B-Instruct	81.3	2894.1 est.	Yes	64.5 est.	No	B200 192GB SXM
qwen35b-a3b-fable-sft-abliterated	34.7	2894.1 est.	Yes	325.2 est.	No	B200 192GB SXM
Llama-3.2-3B-Instruct	3.2	2713.2 est.	Yes	302.3 est.	Yes	B200 192GB SXM
gpt-oss-20b	21.5	2411.7 est.	Yes	268.7 est.	Yes	B200 192GB SXM
gemma-4-26B-A4B-it	25.2	2284.8 est.	Yes	254.5 est.	Yes	B200 192GB SXM

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

	B200 192GB SXM	V100 16GB PCIe
Catalog models that fit	61 of 73	25 of 73
Largest exclusive fit	MiMo-V2.5 (310.0B)	—

B200 192GB SXM details → · V100 16GB PCIe details →