The first benchmarks of AMD's Instinct MI300X 192 GB GPU have been published in Geekbench OpenCL benchmark and it's quite the beast.
AMD Instinct MI300X 192 GB GPU Takes The OpenCL Benchmark Crown, Faster Than NVIDIA's Flagship RTX 4090
The AMD Instinct MI300X is primarily a Data Center GPU designed to meet today's AI requirements. Its main competition is the Hopper AI GPU family from NVIDIA but it looks like someone tested the GPU in the Geekbench OpenCL benchmark and it completely obliterates the entire GPU ladder, taking the top spot with ease.
Before we get into the benchmark itself, we have to talk about the AMD Instinct MI300X specs. Based on the CDNA 3 architecture, the MI300X is an engineering marvel that goes heavy into the chaplet packaging design with 153 billion transistors. There are a total of 28 dies on the chip itself which include eight HBM3 packages for up to 192 GB memory capacities, the highest of all Data Center GPUs that are available today.
On the GPU side, the Instinct MI300X packs a total of 304 compute units with a total of 19,456 cores. The chip features a 5.3 TB/s VRAM bandwidth and a fast 896 GB/s of Infinity Fabric interconnect speeds. As for power consumption, the MI300X has a rated TDP of 750W which is almost twice the figure of NVIDIA's RTX 4090 which is rated at 450W.
So in terms of performance, the AMD Instinct MI300X scored a very impressive 379,660 score in the Geekbench 6.3.0 OpenCL test. The chip was running on an AMD EPYC setup comprising two 9754 CPUs with 128 cores each. The system itself had 3 TB of system memory. If we are to compare this result with today's gaming offerings, the fastest GPU on the planet, the RTX 4090, scores 319,697 points while AMD's flagship, the 7900 XTX, scores 207,354 points.
Geekbench OpenCL (Higher is Better)
This goes off to show just how fast today's modern Data Center and AI GPUs are but due to their high power and cooling requirements plus the insanely high prices, we aren't going to see them on standard PC platforms. For example, a single RTX 4090 sells for around $1500 to $2000 US while a single MI300X chip sells for around $15,000 US.
The other reason why we won't see them on such platforms is because they feature specific hardware designs and accelerators that make no sense for gaming and traditional PC applications. Plus the driver work isn't there either. These chips are specifically made to meet the demands of Data Centers, AI, HPC, and Cloud servers. AMD has already unveiled its next-gen Instinct roadmap with the MI350 and MI400 series.
News Source: Benchleaks
About the author: A Software Engineer by training and a PC enthusiast by passion, Hassan Mujtaba serves as Wccftech's Senior Editor for hardware section. With years of experience in the industry, he specializes in deep-dive technical analysis of next-generation CPU and GPU architectures, motherboards, and cooling solutions. His work involves not only breaking news on upcoming technologies but also extensive hands-on reviews and benchmarking.
Follow Wccftech on Google to get more of our news coverage in your feeds.
