Search

Showing top 28 results for "Model performance updates"

tomshardware.com › pc-components › gpus

Nvidia details Rubin architectural optimizations for inference – improvements target better performance and efficiency from the GPU to the rack

… Softmax on Rubin gets up to a 4X boost versus Blackwell Rubin also focuses on improving the performance of the attention mechanism that’s foundational to transformer-based LLMs More advanced models now support context lengths of up to a million tokens, and quickly performing attention calculations … …

Jul 21, 2026 · Jeffrey Kampman
tomshardware.com › tech-industry › artificial-intelligence

Huawei-led team claims it post-trained DeepSeek's 1.6-trillion-parameter model — 1,000 Ascend 910C chips used in training

… Back in August, it was reported that DeepSeek couldn’t complete a single successful training run for its R2 model in Ascend chips, even with Huawei engineers on site, blaming unstable performance, slow chip-to-chip interconnects, and gaps in Huawei's CANN software stack, its substitute for Nvidia's… …

Jun 6, 2026 · Luke James