← AI Hashrate

RTX 4080 Super 16GB vs RTX 5080 16GB for Local LLMs

The RTX 5080 16GB has about 30% more memory bandwidth (960 vs 736 GB/s). Batch-1 decode tracks bandwidth closely, so a model that fits both cards should decode roughly 30% faster on the RTX 5080 16GB. Both fit the same number of catalog models at Q4/8K (25 of 61; 16 vs 16 GB VRAM). MSRP is $999 for the RTX 4080 Super 16GB vs $999 for the RTX 5080 16GB. 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 (44 shared models), the RTX 5080 16GB posts the higher tok/s in 44 rows vs 0. Methodology.

Spec comparison

RTX 4080 Super 16GBRTX 5080 16GB
VendorNVIDIANVIDIA
VRAM16 GB16 GB
Memory typeGDDR6XGDDR7
Memory bandwidth736 GB/s960 GB/s
TDP320 W360 W
MSRP (list)$999$999
BuyASUS ProArt GeForce RTX 4080 Super OC (affiliate) · More listingsNVIDIA GeForce RTX 5080 Founders Edition (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

ModelParams (B)RTX 4080 Super 16GB tok/sRTX 4080 Super 16GB fitsRTX 5080 16GB tok/sRTX 5080 16GB fitsWinner
Qwen3-0.6B0.8585.5 est.Yes763.6 est.YesRTX 5080 16GB
gemma-3-1b-it1.0468.4 est.Yes610.9 est.YesRTX 5080 16GB
Llama-3.2-1B-Instruct1.2390.3 est.Yes509.1 est.YesRTX 5080 16GB
Qwen3-30B-A3B30.5338.0 est.No440.9 est.NoRTX 5080 16GB
Qwen3-Coder-30B-A3B-Instruct30.5338.0 est.No440.9 est.NoRTX 5080 16GB
Qwen3-30B-A3B-Instruct-250730.5338.0 est.No440.9 est.NoRTX 5080 16GB
GLM-4.7-Flash31.2324.5 est.No423.2 est.NoRTX 5080 16GB
Qwen3-1.7B2.0234.2 est.Yes305.5 est.YesRTX 5080 16GB
gpt-oss-20b21.5186.5 measuredYes204.9 measuredYesRTX 5080 16GB
Llama-3.2-3B-Instruct3.2146.4 est.Yes190.9 est.YesRTX 5080 16GB
Mistral-7B-Instruct-v0.37.265.0 measuredYes163.6 measuredYesRTX 5080 16GB
Phi-4-mini-instruct3.8123.3 est.Yes160.8 est.YesRTX 5080 16GB
Phi-4-mini-reasoning3.8123.3 est.Yes160.8 est.YesRTX 5080 16GB
gemma-3-4b-it4.3120.1 est.Yes156.6 est.YesRTX 5080 16GB
Qwen3-4B-Instruct-25074.0117.1 est.Yes152.7 est.YesRTX 5080 16GB

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 16GBRTX 5080 16GB
Catalog models that fit25 of 6125 of 61
Largest exclusive fit——

RTX 4080 Super 16GB details → · RTX 5080 16GB details →