# Nvidia GeForce RTX 5090 and 5080: Recent Leaks Unveil Possible Specifications but Leave Unanswered Questions
As excitement grows for Nvidia’s upcoming GPUs, fresh leaks have emerged, revealing potential specifications for the forthcoming **GeForce RTX 5090** and **RTX 5080**. While the official **price** and **release date** are still unclear, the disclosed information offers a preview of the performance potential of these eagerly awaited graphics cards. Nonetheless, the details have sparked concerns, especially regarding power use and memory setups.
## The Leak: Current Insights
Respected leaker **kopite7kimi** has revealed comprehensive specifications for both the RTX 5090 and RTX 5080 on X (formerly Twitter). These leaks illuminate the **Blackwell architecture** that will drive Nvidia’s next flagship GPUs, providing enthusiasts with a more refined expectation.
### RTX 5090 Specifications
The leaks indicate that the **RTX 5090** will utilize the **GB202-300-A1** chip, featuring a remarkable **21,760 CUDA cores**. This represents a notable increase from the RTX 4090, which has 16,384 CUDA cores. The memory setup is also significantly enhanced, incorporating **32 GB of GDDR7 memory** linked through a **512-bit interface**. This arrangement promises a considerable boost in memory bandwidth, potentially achieving **2,048 GB/s** at 32 Gbps, in contrast to the RTX 4090’s 1,008 GB/s.
The most surprising element of the RTX 5090 is its **power consumption**, which is projected to reach an astonishing **600 watts**. Despite this elevated power requirement, Nvidia is reportedly implementing a **2-slot cooler**, implying an updated cooling solution to handle the increased thermal output.
### RTX 5080 Specifications
Conversely, the **RTX 5080** is set to incorporate the **GB203-400-A1** chip with **10,752 CUDA cores**—a slight enhancement over the RTX 4080’s 9,728 CUDA cores. The memory configuration includes **16 GB of GDDR7 memory** with a **256-bit interface**, supporting a potential bandwidth of **1,024 GB/s** at 32 Gbps. The power consumption for the RTX 5080 is anticipated to be approximately **400 watts**.
While these specifications are impressive, the memory configuration of the RTX 5080 has ignited discussions. Numerous enthusiasts were hoping for more memory in Nvidia’s second-tier GPU, particularly in light of the growing requirements of contemporary games and applications. This has prompted some to compare it with the controversial launch of the initial RTX 4080, which faced criticism for its limited 12 GB of VRAM. Nvidia eventually rebranded that model to the **RTX 4070 Ti** after encountering backlash.
## Comparison with Prior Generations
To comprehend the importance of these leaks more clearly, here’s a comparison between the leaked specifications of the RTX 5090 and RTX 5080 and their predecessors, the RTX 4090 and RTX 4080:
| **Specification** | **RTX 5090** | **RTX 4090** | **RTX 5080** | **RTX 4080** |
|—————————|————————|————————|————————|————————|
| **Chip** | GB202-300-A1 | AD102-300-A1 | GB203-400-A1 | AD103-300-A1 |
| **CUDA Cores** | 21,760 | 16,384 | 10,752 | 9,728 |
| **Memory Size/Type** | 32 GB GDDR7 | 24 GB GDDR6X | 16 GB GDDR7 | 16 GB GDDR6X |
| **Memory Interface** | 512-bit | 384-bit | 256-bit | 256-bit |
| **Bandwidth** | 2,048 GB/s (32 Gbps) | 1,008 GB/s | 1,024 GB/s (32 Gbps) | 717 GB/s |
| **Power Consumption** | 600 watts | 450 watts | 400 watts | 320 watts |
| **MSRP at Release** | – | €1,949 | – | €1,469 |
### Key Takeaways:
– **Performance Increase**: The RTX 5090 exhibits a significant rise in CUDA cores and memory bandwidth compared to the RTX 4090, indicating a considerable performance enhancement.
– **Power Requirements**: Both the RTX 5090 and RTX 5080 present higher power consumption than their predecessors, which could result in