- Asus rtx 5080 now works as Hardware AI and a surprisingly great SSD bay
- Pcie lanes are no longer just for GPU, thanks to the Asus hybrid hardware approach
- Asus is developing multi-lm so that IA developers can download models directly to the SSD on board
The unusual GPU of GeForce RTX 5080, first seen a few weeks ago with a SSD slot, is no longer just a rarity of hardware.
Asus North America has now confirmed that the SSD Proartx 5080 edition is a real product.
However unusual it may seem, the decision to combine high -performance graphics processing with solid state storage is part of a broader plan that involves artificial intelligence and development of local models.
A storage design gpu hybrid aimed at the development of AI
Although Asus remains with tight lips on a release date and shipping configuration, some technical details have begun to emerge, along with some unanswered questions.
One of the biggest surprises is the 90 -degree rotated GPU design: ASUS modified the design of the printed circuit plate (PCB) so that the GPU is mounted perpendicular to the usual orientation.
This is not a configuration found in any other RTX 5080 model, which suggests a personalized approach that accommodates the SSD and improves the air flow.
Asus states that this design helps the SSD to benefit from “Airflow provided by card fans”, although the SSD is not in direct contact with the main thermal dissipator.
Instead, it cools through a secondary thermal dissipator and a passive air flow, according to the reports, keeps the SSD approximately 10 degrees Celsius cooler than a typical m.2 unit mounted on a motherboard.
SSD integration also exploits PCIE fork, which means that the unused PCI Express lanes are reassigned to admit the unit.
The company has previously applied this approach to the RTX 4060 series cards, which did not completely saturate the PCIE interface, leaving enough bandwidth for an M.2 slot.
Now, with the RTX 5080 Proart edition, the idea is refined to adapt to high -end IA workflows, although it introduces compensation, the GPU may not have access to all PCIE lanes, which potentially limits maximum performance under certain workloads.
Perhaps the most significant revelation lies in ASUS software ambitions: the company is developing a tool called Multi -LM, aimed at developers working with large language models.
Asus says that this tool “will allow the developers to download models directly to storage, which potentially allows the development of local models without the need for constant memory exchanges to slower system units.”
This will be of interest to those looking for the best LLM for coding, and positions the RTX 5080 SSD edition as more than a game card; It could become one of the best AI tools for local inference and the creation of prototypes.
Asus has only launched a single card render and has not confirmed the metrics or performance availability.
The question of whether this concept offers significant real -world profits or simply adds complexity remains unanswered.
Through Videocardz
