Skip to main content
  1. Home
  2. Computing
  3. News

The RTX 5090 will reportedly require 600 watts of power

Add as a preferred source on Google
The back of the Nvidia RTX 4080 Super graphics card.
Jacob Roach / Digital Trends

Rumors have been circulating about the RTX 5090 for a while, but we’re finally getting a clearer picture of how Nvidia’s flagship RTX 50-series GPU is shaping up. Well-known hardware leaker Kopite7kimi is claiming the RTX 5090, which seems destined to earn a spot among the best graphics cards, will come with an obscene power requirement of 600 watts.

The leaker shared specs for the RTX 5090 and RTX 5080 on X (formerly Twitter). We’ve seen vague claims from Kopite7kimi in the past concerning RTX 50-series graphics cards, but this is the first time the leaker has shared full specs. According to the leak, the RTX 5090 will require 600W of power and come with a staggering 32GB of GDDR7 memory across a 512-bit bus.

Recommended Videos

GeForce RTX 5090
PG144/145-SKU30
GB202-300-A1
21760FP32
512-bit GDDR7 32G
600W

— kopite7kimi (@kopite7kimi) September 26, 2024

We’ve heard murmurs in the past few months that the RTX 5090 could reach as high as 600W, but this is the first time it’s been brought up alongside other specs. The RTX 4090, by comparison, requires 450W of power. However, the 16-pin power connection the RTX 4090 uses is capable of 600W, and it seems Nvidia could max out that connection with the next-gen RTX 5090. Despite the increased power demands, Kopite7kimi still claims the RTX 5090 will use a dual-slot cooler.

The leaker also shared details about the RTX 5080, claiming the GPU will require 400W and come with 16GB of GDDR7 memory on a 256-bit bus. That’s not a massive improvement in specs compared to the RTX 4080 Super, suggesting Nvidia will keep a large gap in performance between the RTX 5090 and RTX 5080 as it did with its RTX 40-series graphics cards.

GeForce RTX 5080
PG144/147-SKU45
GB203-400-A1
10752FP32
256-bit GDDR7 16G
400W

— kopite7kimi (@kopite7kimi) September 26, 2024

Nvidia has yet to confirm that the RTX 5090 and RTX 5080 even exist, though we already know quite a bit about the next-gen Blackwell architecture the GPUs will use. Nvidia has already introduced this architecture to the data center, despite the fact that it was forced to delay the release of its enterprise Blackwell GPUs due to a design flaw.

It’s not clear when Nvidia will introduce its next-gen GPUs to gaming PCs. Following Nvidia’s normal release cadence, we would expect to see the GPUs in October or November — however, given the delay in the enterprise sector, it’s possible the GPUs could get pushed back to 2025.

Jacob Roach
Former Lead Reporter, PC Hardware
Jacob Roach is the lead reporter for PC hardware at Digital Trends. In addition to covering the latest PC components, from…
Windows 11 is about to turn on a setting that could hurt gaming performance
Microsoft will start enabling Memory Integrity on more Windows 11 PCs in October
A gaming PC with RGB synced lights running Apex Legends.

Microsoft is preparing to enable Memory Integrity on more Windows 11 PCs starting in October. The security feature is meant to protect systems from malicious code, but there is one reason gamers may want to keep an eye on it.

Microsoft has previously acknowledged that Memory Integrity can affect gaming performance on some Windows 11 systems. Back in 2022, the company even published instructions explaining how gamers could temporarily disable Memory Integrity and Virtual Machine Platform if they were causing problems.

Read more
AI chatbots will agree and misinform if you just put a little pressure, warns research
ChatGPT 3.5 proved the most vulnerable when false claims were repeated over and over
Claude AI on an iPhone.

AI chatbots have a well-documented habit of hallucinating information and sometimes agreeing with users even when they are wrong. A new study suggests that simply refusing to take no for an answer can make the problem worse.

Researchers from the University of Arizona tested seven AI models, including GPT-3.5, GPT-4o, GPT-4o-mini, Claude 3.5 Sonnet, Gemini 1.5 Pro, Llama 3 70B, and DeepSeek-R1. Instead of judging them from a single response, researchers kept conversations going while repeatedly feeding the models information they knew was false.

Read more
Google brings its best AI music model Lyria 3.5 to the Gemini app
Developers get full API access through Google AI Studio.
Google-gemini-lyria-3.5

Google has added Lyria 3.5, its most advanced music generation model yet, to the Gemini app. Previously available through Google's AI filmmaking tool Flow, the model is now rolling out to all Gemini users, making it easier to generate polished songs, instrumentals, and soundtracks from simple text prompts or even photos.

Lyria 3.5 makes AI-generated music sound more natural

Read more