Nvidia has yet to confirm the existence of entry-level GPUs in its latest RTX 50 series. However, recent leaks have revealed the anticipated specifications of the upcoming RTX 5050, 5060, and 5060 Ti. These GPUs are expected to be officially announced soon, as reports from last week indicated that Nvidia plans to unveil additional RTX 50 series models within the next 10 days.
As for the leaked specifications, @kopite7kimi, a fairly accurate leaker, has shared important information about the upcoming GPUs. As per his posts on X, Nvidia’s RTX 5060 and RTX 5060 Ti will utilize the GB206-300 GPU die, while the RTX 5050 will be built around the GB207-300 chip. The RTX 5060 Ti, positioned as the more powerful of the two 60-class cards, is expected to feature 4,608 CUDA cores—just a modest 5.7% increase over its predecessor, the RTX 4060 Ti.
Recommended Videos
GeForce RTX 5060 TiPG152-SKU10/15GB206-300-A14608FP32128-bit GDDR7 8/16G180W
— kopite7kimi (@kopite7kimi) March 9, 2025
Get your weekly teardown of the tech behind PC gaming
ReSpec
Subscribe
Check your inbox!
Privacy Policy
The GPU will come in two memory configurations: 8GB and 16GB of GDDR7, both operating on a 128-bit memory bus with a 180W TDP. The 16GB variant is set to launch first, hitting shelves by the end of March, while the 8GB model will follow in April alongside the RTX 5060.
GeForce RTX 5060PG152-SKU25GB206-250-A13840FP32128-bit GDDR7 8G150W
— kopite7kimi (@kopite7kimi) March 10, 2025
The RTX 5060 itself will be available in a single 8GB GDDR7 variant, maintaining the same 128-bit memory bus. It will feature 3,840 CUDA cores—a notable 22.22% increase over the RTX 4060. With a 150W TDP, this GPU is expected to be a solid choice for small form factor PC builds.
GeForce RTX 5050PG152-SKU50GB207-300-A12560FP32128-bit GDDR6 8G130W
— kopite7kimi (@kopite7kimi) March 9, 2025
As for the RTX 5050, this marks the return of Nvidia’s 50-class GPU lineup, last seen with the RTX 3050. While the RTX 5060 and 5060 Ti will benefit from a 128-bit memory interface with a bandwidth of 448 GB/s—roughly 55% higher than the RTX 4060 Ti—the RTX 5050 will remain on GDDR6 memory. It is expected to use the PG152-SKU10/15 board with a GB206-300-A1 GPU die, offering 2,560 CUDA cores and a TDP of 130W.
The RTX 50 series, including the RTX 5090, 5080, 5070 Ti, and 5070, is off to a sluggish start, primarily due to limited availability and inflated pricing. Concerns have also arisen regarding the flagship RTX 5090’s incompatibility with older 12VHPWR power connectors, which has led to reports of cable and adapter melting under high power loads. Additionally, Nvidia is facing issues with certain RTX 50 graphics cards having missing ROPs (Render Output Units), resulting in a reduction in overall performance.
Editors’ Recommendations
-
PSA: If you bought an Nvidia RTX 50, check it for problems
-
Nvidia’s RTX 5060 Ti might give AMD an unexpected edge
-
RTX 50 series no longer supports one of Nvidia’s oldest GPU technologies
-
The RTX 5070 Ti may continue Nvidia’s disappointing streak
-
RTX 5090 GPU shortage could soon end if Nvidia adopts this strategy
