NVIDIA GeForce RTX 40 Series May Include a Flagship Model with 900-Watt TGP, 48 GB of GDDR6X Memory, and Dual 16-Pin Power Connectors

The FPS Review may receive a commission if you purchase something after clicking a link in this article.


The power demands of NVIDIA’s endmost next-generation GeForce flagship could turn out to be way, way higher than anyone might’ve imagined. That’s according to the latest report from leaker kopite7kimi, who is now claiming that green team is mulling a graphics card that leverages the full AD102 GPU and features a relatively insane 900-watt TGP—300 watts higher than the already uncomfortable number that’s been rumored for the GeForce RTX 3090. This model is said to boast 48 GB of 24 Gbps GDDR6X memory, dual 16-pin power connectors, and, perhaps unsurprisingly, higher clock speeds.

NVIDIA reportedly testing 900W graphics card with full next-gen Ada AD102 GPU (VideoCardz)

It should be noted that multiple sources now claim that the top GeForce RTX 40 SKU that is supposed to launch later this year will be designed for 600W TDP, which could be the next-gen GeForce RTX 4090 SKU. Therefore, the rumored 900W variant is either an RTX 4090 Ti that is supposed to launch later or a side-project that might at some point end up as a real product. One also cannot rule out that NVIDIA will be bringing back its TITAN series, because the leaker also claims that it will feature 48 GB of 24Gbps memory.

Kopite7kimi has also shared some of the alleged specifications for NVIDIA’s upcoming GeForce RTX 4080 (AD103) and GeForce RTX 4070 (AD104) graphics cards. The former is said to use the AD103 GPU and feature 16 GB of GDDR6X memory and a TGP that’s similar to the GA102 models, while the latter is said to use the AD104 GPU and feature 12 GB of GDDR6 memory and a TGP of 300 watts.

Join the discussion in our forums...

Tsing Mui
News poster at The FPS Review.

Recent News