Nvidia is reportedly preparing a major upgrade to its laptop GPU lineup. According to an apparent leak shared by security researcher Dominic Alvieri, the upcoming RTX 50 series (codenamed “GN22”) will feature six variants, surpassing the five options currently available in the RTX 40 series.
If accurate, the most notable improvement in the updated specs is the increase in video random access memory (VRAM). The standard for the RTX 50 series is 8GB of VRAM, a step up from the 6GB standard for current laptop GPUs.
The lineup will reportedly include three models with 8GB of VRAM, two with a powerful 16GB, and one with 12GB.
These improvements in VRAM allow new GPUs to run more artificial intelligence (AI) models that often require significant memory resources. For example, Stable Diffusion 1.5 requires at least 4GB of VRAM to run properly. SDXL requires at least 6GB to run.
Stability AI used a GPU with 24GB of VRAM when previewing the upcoming SD3 model, which is much more capable and powerful. Fortunately, various versions will be available, covering 800 million to 8 billion parameters, and various configurations, such as excluding the 4.7 billion parameter T5 text encoder, are not very difficult.
According to the leak, the RTX 50 series will utilize faster GDDR7 memory to further improve AI-related tasks and gaming performance.
A leaked roadmap suggests that Nvidia plans to continue producing some older GPUs, such as the RTX 4050 and RTX 3050, with 4GB and 6GB VRAM to meet market demand for cheaper hardware. For gaming, a 6GB card is still a solid option because VRAM isn’t as important as clock speed and processor or core count.
No specific timeline was provided, but this GPU will reportedly be released in 2025. There have been conflicting rumors, but a CES 2025 reveal is possible.
But buyers looking for a cheap card may be running out of time. Nvidia’s main chip supplier, Taiwan Semiconductor Manufacturing Company (TSMC), is said to be considering raising prices as its own costs increase.
“We expect customers to share the higher costs with us and we have already started discussions with customers,” Wei said, according to Nikkei Asia.
According to the report, Wei recently referred to Nvidia CEO Jensen Huang as a “three trillion person” and referenced Nvidia’s soaring market value, which surpassed $3 trillion last week.
TSMC produces approximately 90% of the world’s most advanced processor chips and plays a critical role in Nvidia’s success.
Despite the potential for increased production costs, Huang expressed support for TSMC’s argument, saying the price increases are “consistent with the value they provide.” Nikkei Asia It has been reported.
Morgan Stanley analysts suggested that if Nvidia accepts TSMC’s price increase, other major AI players may also increase their prices accordingly. They estimate that Nvidia will account for 10% of TSMC’s revenue this year.
generally intelligent newsletter
A weekly AI journey explained by Gen, a generative AI model.
