The PyTorch Foundation recently released PyTorch version 2.5, which contains support for Intel GPUs. The release also includes several performance enhancements, such as the FlexAttention API, TorchInd ...

Intel is calling this concept Project Battlematrix. For an eight-GPU system, you're looking at about 1.5 petaOPS of dense INT8 performance, 192GB of vRAM, and 3.6TB/s of aggregate memory bandwidth.

Intel just put out a blog post detailing how it's possible to run Meta AI's Llama 2 large language model on one of its Arc A770 graphics cards.

As part of the PyTorch 2.0 compilation stack, the TorchInductor CPU backend optimization by Intel Extension for PyTorch and PyTorch ATen CPU achieved up to 1.7 times faster FP32 inference ...

Intel announces the 2023 release of Intel oneAPI tools – available in Intel Developer Cloud and rolling out through regular distribution channels.

Intel says that version 32.0.101.6734 of its graphics driver can speed up average frame rates in some games by around 10 percent and can speed up "1 percent low FPS" (that is, for any given frames ...

Intel can sell all of Gaudi 3 and Falcon Shores that it can make in the short term. Now is the time to hustle. The timing for Gaudi 3 and Falcon Shores could have been better, but any delays in ramps ...

Intel fellow Tom Petersen says that the cascade of ongoing Intel GPU driver releases means that the buggy launch of its first-generation Alchemist discrete GPUs are a problem no longer.

Intel says that its new Gaudi 3 AI accelerator offers up to 1856 BF16/FP8 matrix TFLOPS as well as up to 28.7 BF16 vector TFLOPS at around 600W TDP, which when compared to the NVIDIA H100 AI GPU ...

Performance-wise, Intel said a 2-Stack Ponte Vecchio GPU configuration on a single OAM is capable of up to 52 TFLOPs of FP64/FP32 compute, 419 TFLOPs of TF32 (XMX Float 32), 839 TFLOPs of BF16 ...