Nvidia RTX 5070 mobile GPU looks set to get more VRAM despite global memory crisis — 8GB could be boosted to 12GB, according to new listings

2 hours ago 8
Nvidia Blackwell silicon (Image credit: Nvidia)

The GeForce RTX 5070 is one of the best graphics cards for gaming, and the mobile version isn't far behind. Nvidia's initial decision to reduce the memory capacity of the mobile variant disappointed many gamers and raised concerns amid a landscape where many games use more than 8GB. However, multiple manufacturer listings (via Huang514613) indicate an unannounced GeForce RTX 5070 Mobile SKU with 12GB of GDDR7, matching the desktop version and ensuring mobile gamers have access to the same capacity.

Go deeper with TH Premium: GPUs

While the GeForce RTX 5070 Mobile will never be as fast as the desktop counterpart, its restored 12GB memory capacity is a major win for laptop gamers. Tech enthusiast Huang514613 recently spotted Lenovo’s new Panther Lake-powered Yoga Pro 7i Aura Edition (12IPH11) officially listed with "Up to NVIDIA GeForce RTX 5070 GPU with 12GB GDDR7 and DLSS 4." Asus is also on board, advertising its ROG Zephyrus G14 2026 (GU405AP-SY022X) with a "NVIDIA GeForce RTX 5070 Laptop GPU 12GB GDDR7." When two major manufacturers independently confirm these specifications, it’s unlikely that this is a listing error.

Newegg listings for the Asus Rog Strix G16 and G18, respectively, also list Nvidia GeForce RTX 5070 Laptop GPU 12GB GDDR7 under graphics card, while confusingly listing 8GB GDDR7 in the video memory spec. Best Buy's HP Omen listing also notes an RTX 5070 laptop GPU with 12GB of VRAM.

Yoga Pro
(Image credit: Tom's Hardware)

GeForce RTX 5070 Mobile 12GB Specifications

Swipe to scroll horizontally

Graphics Card

GeForce RTX 5070 Mobile 12GB*

Radeon RX 7900M

Radeon RX 7800M

GeForce RTX 5070 Mobile 8GB

Architecture

Blackwell

RDNA 3

RDNA 3

Blackwell

GPU

GB206

Navi 31

Navi 32

GB206

Shaders

4,608

4,608

3,840

4,608

Base Clock (MHz)

907

1,825

1,295

907

Boost Clock (MHz)

1,425

2,090

2,335

1,425

Memory Capacity

12GB GDDR7

16GB GDDR6

12GB GDDR6

8GB GDDR7

Memory Clock (Gbps)

24

18

18

24

Memory Interface

128-bit

256-bit

192-bit

128-bit

Memory Bandwidth (GB/s)

384

576

432

384

TDP (W)

50

180

180

50

*Specifications are unconfirmed by Nvidia.

Neither Lenovo nor Asus has disclosed detailed specifications about the GeForce RTX 5070 Mobile 12GB beyond the spotlight-grabbing memory capacity boost. While specifics are still under wraps, it’s reasonable to anticipate that most core specifications will remain unchanged. The notable jump in memory capacity may result in a modest increase in TDP.

Although RDNA 4 has been on the market for a year now, AMD has yet to launch any mobile variants. The market is still awaiting the arrival of the Radeon RX 9000M series, reportedly spearheaded by the RX 9080M and followed by the RX 9070M XT, RX 9070M, and RX 9060M. In the meantime, we only have RDNA 3 for comparison. With its recovered memory, the GeForce RTX 5070 Mobile 12GB matched the Radeon RX 7800M in capacity but still trails the Radeon RX 7900M.

Get Tom's Hardware's best news and in-depth reviews, straight to your inbox.

8GB graphics cards are already struggling with modern games, so it's not a coincidence that there's been a slow but steady transition to 12GB in recent years. 8GB accounts for almost one-third (29.37%) of the Steam gamer population. However, 12GB (15.22%) and 16GB (18.26%) will continue to grow.

Given that we're still in the thick of the memory shortage, it came as a surprise that Nvidia would give the GeForce RTX 5070 Mobile a memory upgrade. Then again, there is always a motive behind every decision, and Nvidia is probably preparing for mobile RDNA 4.

Google Preferred Source

Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds.

Zhiye Liu is a news editor, memory reviewer, and SSD tester at Tom’s Hardware. Although he loves everything that’s hardware, he has a soft spot for CPUs, GPUs, and RAM.

Read Entire Article