NVIDIA A100 Accelerator Based On Next-Gen Ampere Architecture With 6912 CUDA Cores And 40GB HBM2 Memory Benchmark Results Out

A record-breaking Graphics Card benchmark has been reportedly set by an NVIDIA Ampere-based GPU. The NVIDIA A100 Accelerator, a purpose-built Graphics Card, appears to be the fastest GPU ever recorded. It is, however, important to note that there are some limitations to the claim.

A compute accelerator from NVIDIA, the A100 Ampere Tensor Core GPU is now the fastest GPU ever recorded, claims an openly shared benchmark result offered by Jules Urbach, the CEO of OTOY (a company specializing in holographic rendering in the cloud). The purpose-specific GPU reportedly produced the best ever results in the OctaneBench bench-marking tool.

NVIDIA A100 Accelerator OctaneBench Results Prove It Is The Best Performing Graphics Card:

Jules Urbach, the CEO of cloud graphics firm OTOY, took to Twitter to share the OctaneBench benchmark results. Incidentally, OctaneBench is developed by OTOY. The benchmarking platform uses the company’s OctaneRender software. The company claims to provide ‘a level playing field by making sure that everybody uses the same version and the same scenes and settings.’ Interestingly, the OctaneRender was the first commercially available ray-tracing tester that is capable of fully utilizing the NVIDIA GPU


OctaneRender is specifically designed for NVIDIA GPUs. It is specific to NVIDIA Graphics Card because of the way it utilizes CUDA technology and ray-tracing tech to push a GPU to its limits of performance. But this also means there won’t be any comparison to AMD Big Navi or Arcturus-based graphics cards.

Coming to benchmark results, according to an image shared by Urbach, the A100 recorded a total score of 446.03. The score was determined by compiling the results from twelve separate tests. The OTOY CEO claims the results place the A100 at 43 percent faster than existing Turing cards. However, Urbach did not mention which Turing-based graphics card accelerator was the results being compared with. It is possible he might be referring to the best scoring GPUs from NVIDIA’s previous generation.

NVIDIA A100 Accelerator OctaneBench Results And Specifications:

As mentioned there’s no information about which result is being compared to A100. However, the fastest Turing-based graphics card in OctaneBench is listed as GRID RTX 8000, which scored 328 points. The rest of the comparative benchmark results indicate the Volta-based Tesla V100, TITAN V, and Quadro GV100 are still a worthy competition to Ampere. The A100, however, shows about 11 to 33 percent performance boost compared to the previous generations.

[Image Credit: Otoy via VideoCardz]
The A100 features NVIDIA’s first 7nm GPU, the GA100. This GPU is equipped with 6912 CUDA cores and 40GB of HBM2 memory. This is also the first card featuring PCIe 4.0 interface or SXM4. The features of the card vary depending on the requirement and purpose.

The NVIDIA A100 GPU Accelerator is not intended for typical consumers or desktop computer systems running typical consumer-facing operating systems. Nonetheless, the results do offer a glimpse of what NVIDIA GPU users can expect from future high-performance NVIDIA products. Moreover, the A100 is the first, and so far, the only, Ampere-based graphics card to surface. Additionally, NVIDIA has confirmed the immediate availability of the A100 for DGX100 systems.

Alap Naik Desai
A B.Tech Plastics (UDCT) and a Windows enthusiast. Optimizing the OS, exploring software, searching and deploying solutions to strange and weird issues is Alap's main interest.