Chinese OEM lists modified RTX 5090 with 96GB VRAM on Alibaba for $3,888
Shenzhen Suqiao Intelligent Technology, an established Chinese OEM, has listed a modified Nvidia GeForce RTX 5090 equipped with 96GB of memory on Alibaba for $3,888. This configuration offers three times the VRAM of the standard consumer card at a price significantly lower than the U.S. market rate for the original GPU. The modification appears to utilize the GB202 silicon, potentially…
Key points
- Shenzhen Suqiao lists a modified RTX 5090 with 96GB VRAM on Alibaba for $3,888.
- The card likely uses GB202 silicon on a custom PCB, mimicking the RTX Pro 6000 Blackwell memory layout.
- Listing details contain contradictory specs, such as GDDR6X references, raising questions about the card's authenticity.
While the listing raises technical skepticism due to contradictory specifications—such as references to GDDR6X memory and speeds inconsistent with GDDR7—the existence of such a card is technically plausible given the RTX Pro 6000 Blackwell’s 96GB configuration. The emergence of these high-memory consumer cards is driven by the AI boom, creating demand for affordable hardware capable of running large language models locally. If authentic, this card would offer substantial value for AI enthusiasts, though its availability and reliability remain unverified due to the gray-market nature of the sale and potential firmware modifications required to enable the extra memory.
The story so far
3 episodes →- Chinese OEM lists modified RTX 5090 with 96GB VRAM on Alibaba for $3,888this story
China-modified Nvidia RTX 5090 with massive 96GB of memory appears on Alibaba for less than $4,000 — 3x more VRAM at 65% the cost of the original
Tom's Hardware · 11 September 2026
Loading the full article…
This text was published by Tom's Hardware and written by Zhiye Liu. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Hardware & Compute
All →- Applied Materials and Intel expand AI chipmaking collaboration at epic Center · 1 src
- Xeal to deploy 100,000 Nvidia GPUs in EV charging pods across US · 1 src
- Meta’s Muse boosts AMD stock as AI agents shift workload to CPUs · 1 src
- Nokia CEO says data center construction could double without supply constraints · 1 src
- Ghost launches $3,499 AI agent computer with $11M seed · 1 src
Comments
via GitHub Discussions