← AI Hashrate

H200 141GB SXM5 vs RTX 3080 10GB for Local LLMs

The H200 141GB SXM5 has about 532% more memory bandwidth (4,800 vs 760 GB/s). Batch-1 decode tracks bandwidth closely, so a model that fits both cards should decode roughly 532% faster on the H200 141GB SXM5. The H200 141GB SXM5 fits more of the catalog at Q4/8K — 48 vs 17 of 61 models (141 vs 10 GB VRAM). Largest model that fits the H200 141GB SXM5 but not the RTX 3080 10GB: MiniMax-M2.7 (228.7B). MSRP is $35,000 for the H200 141GB SXM5 vs $699 for the RTX 3080 10GB. MSRP is a launch list price, not live retail — Amazon prices move; click through for the current price. Not retail-shoppable: H200 141GB SXM5 (datacenter-class; cloud rental is the realistic way to use it). In the Q4/8K head-to-head below (38 shared models), the H200 141GB SXM5 posts the higher tok/s in 38 rows vs 0. Methodology.

Spec comparison

H200 141GB SXM5RTX 3080 10GB
VendorNVIDIANVIDIA
VRAM141 GB10 GB
Memory typeHBM3eGDDR6X
Memory bandwidth4,800 GB/s760 GB/s
TDP700 W320 W
MSRP (list)$35,000$699
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 38 shared models

ModelParams (B)H200 141GB SXM5 tok/sH200 141GB SXM5 fitsRTX 3080 10GB tok/sRTX 3080 10GB fitsWinner
Qwen3-0.6B0.83818.2 est.Yes604.5 est.YesH200 141GB SXM5
gemma-3-1b-it1.03054.5 est.Yes483.6 est.YesH200 141GB SXM5
Llama-3.2-1B-Instruct1.22545.5 est.Yes403.0 est.YesH200 141GB SXM5
Qwen3-1.7B2.01527.3 est.Yes241.8 est.YesH200 141GB SXM5
Qwen3-30B-A3B30.51018.2 est.Yes136.3 est.NoH200 141GB SXM5
Qwen3-Coder-30B-A3B-Instruct30.51018.2 est.Yes136.3 est.NoH200 141GB SXM5
GLM-4.7-Flash31.21018.2 est.Yes130.9 est.NoH200 141GB SXM5
Qwen3-30B-A3B-Instruct-250730.51018.2 est.Yes136.3 est.NoH200 141GB SXM5
Llama-3.2-3B-Instruct3.2954.5 est.Yes151.1 est.YesH200 141GB SXM5
gpt-oss-20b21.5848.5 est.Yes209.1 est.NoH200 141GB SXM5
Phi-4-mini-instruct3.8803.8 est.Yes127.3 est.YesH200 141GB SXM5
Phi-4-mini-reasoning3.8803.8 est.Yes127.3 est.YesH200 141GB SXM5
gemma-3-4b-it4.3783.2 est.Yes124.0 est.YesH200 141GB SXM5
Qwen3-4B-Instruct-25074.0763.6 est.Yes120.9 est.YesH200 141GB SXM5
Qwen3-4B-Thinking-25074.0763.6 est.Yes120.9 est.YesH200 141GB SXM5

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

H200 141GB SXM5RTX 3080 10GB
Catalog models that fit48 of 6117 of 61
Largest exclusive fitMiniMax-M2.7 (228.7B)—

H200 141GB SXM5 details → · RTX 3080 10GB details →