More Watts, BIG Memory – Nvidia GeForce RTX 5080 SUPER Leaks

Nvidia’s RTX 5080 SUPER goes BIG on memory and BIG on power

Specifications for Nvidia’s GeForce RTX 5080 SUPER have leaked, unveiling a HUGE increase in both memory capacity and TGP (Total GPU Power), signalling a substantial increase in GPU performance.

According to @kopite7kimi, Nvidia’s upcoming RTX 5080 SUPER will feature a TGP of 415W and 24GB of 32 Gbps GDDR7 memory. Compared to today’s RTX 5080, Nvidia’s new SUPER variant will feature 50% more VRAM capacity, 6.6% more memory bandwidth, and a higher power budget. With 55 more watts at its disposal, Nvidia’s new SUPER GPU should be able to target higher clock speeds, boosting performance.

Note that Nvidia’s RTX 5080 and RTX 5080 SUPER have the same number of CUDA cores. Both feature 10752 CUDA cores. In a sense, this makes Nvidia’s new SUPER variant an overclocked RTX 5080 model with added memory capacity. Aside from their power budget and memory speeds, the only major difference between these cards is the type of GDDR7 DRAM modules they use. Today’s RTX 5080 utilises 2GB GDDR7 modules, whereas its SUPER counterpart reportedly employs 3GB modules.

If you’re not interested in having access to more VRAM, Nvidia’s RTX 5080 appears to be little more than an overclocked RTX 5080. That said, these specifications could change ahead of Nvidia’s RTX 50 SUPER launch. We expect Nvidia’s RTX 50 SUPER series to launch in early 2026, one year after the release of Nvidia’s standard RTX 50 series GPU models.

You can join the discussion on Nvidia’s leaked GeForce RTX 5080 SUPER GPU specifications on the OC3D Forums.

Mark Campbell

Mark Campbell

A Northern Irish father, husband, and techie that works to turn tea and coffee into articles when he isn’t painting his extensive minis collection or using things to make other things.

Follow Mark Campbell on Twitter
View more about me and my articles.