Nvidia GeForce RTX 50 SUPER “Back on Track” – Leaker claims
Nvidia’s RTX 50 SUPER series GPUs could launch this year
Nvidia has reportedly unshelved its plans for new GeForce RTX 50 “SUPER” series graphics cards, with prominent leaker MEGAsizeGPU stating they are “back on track”. These planned GPUs were reportedly cancelled due to rising memory prices and shortages. Now, it looks like Nvidia has found a way to work around these issues.
Nvidia’s new RTX 50 SUPER lineup reportedly includes new “SUPER” variants of its RTX 5060, RTX 5070, RTX 5070 Ti, and RTX 5080. All of these GPUs will reportedly use 3GB GDDR7 memory chips, replacing the 2GB chips in their non-SUPER counterparts. This will increase their memory capacity by 50%. This means that the RTX 5060 SUPER, RTX 5070 SUPER, RTX 5070 Ti SUPER, and RTX 5080 SUPER will have 12GB, 18GB, 24GB, and 24GB of VRAM, respectively. These GPUs will also feature higher CUDA core counts and/or higher TDPs to increase their compute performance.
RTX 50 Super is back on track. This time includes 5060 12G ( or maybe it will have a new name as 5060 super )
— MEGAsizeGPU (@Zed__Wang) June 5, 2026
RTX SUPER Specifications
MEGAsizeGPU believes that Nvidia plans to release its new “SUPER” series GPUs this year. Specifications remain the same as prior leaks. However, pricing remains unknown.
Earlier this year, Micron called GDDR7 memory capacity a “performance bottleneck” when unveiling its 3GB GDDR7 modules. They can reduce texture pop-in by allowing more texture data to be in memory. It can also allow for more detailed environments with fewer loading transitions. Micron’s post came in February 2026, around the time Nvidia was expected to release its RTX 50 SUPER series GPUs. These plans got shelved due to the ongoing memory supply crisis.
It's back on track, but I don't have a specific time window for the release. But I believe it will be in 2026.
— MEGAsizeGPU (@Zed__Wang) June 5, 2026
It is currently unknown how Nvidia intends to work around the memory supply crisis. As it stands, adding 50% more memory to modern GPUs will ensure a massive price increase for these new “SUPER” GPU models. Has Nvidia found a way to secure its memory at a lower price? Does Nvidia expect memory prices to fall soon? If that is the case, does Nvidia expect the AI bubble to pop soon?
| RTX 5060 | RTX 5060 SUPER |
RTX 5070 | RTX 5070 SUPER (Leaked) |
RTX 5070 Ti | RTX 5070 Ti SUPER (Leaked) |
RTX 5080 | RTX 5080 SUPER (Leaked) |
|
| SM Count | 30 | ??? | 48 | 50 | 70 | 70 | 84 | 84 |
| CUDA Cores | 3840 | ??? | 6144 | 6400 | 8960 | 8960 | 10752 | 10752 |
| Memory Capacity | 8GB | 12GB | 12GB GDDR7 | 18GB GDDR7 | 16GB GDDR7 | 24GB GDDR7 | 16GB GDDR7 | 24GB GDDR7 |
| Memory Speed | 28 Gbps | ?? | 28 Gbps | 28 Gbps | 28 Gbps | 28 Gbps | 30 Gbps | 32 Gbps (maybe higher) |
| Memory Bus Size | 128-bit | 128-bit | 192-bit | 192-bit | 256-bit | 256-bit | 256-bit | 256-bit |
| TGP | 145W | ?? | 250W | 275W | 300W | 350W | 360W | 415W |
You can join the discussion on Nvidia’s RTX 50 SUPER series being “back on track” on the OC3D Forums.
