Mac Pro
PulseAugur coverage of Mac Pro — every cluster mentioning Mac Pro across labs, papers, and developer communities, ranked by signal.
5 day(s) with sentiment data
-
M7 Ultra chip may support 1.5TB RAM, matching 2019 Mac Pro
The upcoming M7 Ultra chip is rumored to support an unprecedented 1.5TB of RAM. This massive memory capacity would bring it in line with the high-end configurations previously offered by the 2019 Mac Pro. This developme…
-
Apple Shelved Two Mac Pro Models, Including Intel Version
Apple Inc. developed two Mac Pro models that were ultimately shelved and never released to the public. One of these unreleased models was based on Intel processors, indicating a potential shift in Apple's strategy or co…
-
Apple reportedly scrapped plans for new Intel Mac Pro model
Apple Inc. reportedly had plans for a new Intel-based Mac Pro model that were ultimately scrapped. This comes after the company announced in March that it was discontinuing the Mac Pro line and had no future hardware pl…
-
Apple reportedly scrapped plans for M2 and M3 Extreme Mac Pro chips
Apple Inc. reportedly developed two high-performance chips, the M2 Extreme and M3 Extreme, intended for the Mac Pro. However, the company ultimately decided against releasing these chips. This information comes from Blo…
-
Apple's M7 Ultra chip targets 1.5TB memory, Blackwell-class AI performance
Apple is reportedly developing its M7 Ultra chip, slated for a 2028 release, with a target of supporting up to 1.5TB of unified memory. This ambitious memory capacity is contingent on the availability of DRAM in the mar…
-
Old Mac Pro repurposed for local LLM tasks with new drivers
An old Mac Pro, originally costing nearly £10,000, is being repurposed for local LLM work thanks to new Linux drivers that enable its D700 GPUs. The machine, equipped with 64GB of RAM and 24 cores, can now run models vi…
-
HeadQ: Model-Visible Distortion and Score-Space Correction for KV-Cache Quantization
Researchers are developing several novel methods to optimize the Key-Value (KV) cache in large language models, which is a major bottleneck for long-context processing. These approaches include training models to inhere…