Yeah, I priced an M3 with 192GB out of curiosity just now and my eyes bulged quite a bit
AI HARDWARE
-
M4 Mac Desktops to Support 512GB Unified Memory
By
–
Wow, might be worth waiting for the M4 then: https://
macrumors.com/2024/04/11/m4-
ai-chips-late-2024/
… "M4 versions of the Mac desktops could support as much as 512GB Unified Memory, which would be a marked jump over the current 192GB limit." -
AI Pin Review: $699 Device Criticized as Broken and Unfinished
By
–
“The AI Pin is an interesting idea that is so thoroughly unfinished and so totally broken in so many unacceptable ways that I can’t think of anyone to whom I’d recommend spending the $699 for the device and the $24 monthly subscription.” https://
theverge.com/24126502/human
e-ai-pin-review
… -
Bixtral deployment errors on A100 GPUs HuggingFace endpoints
By
–
Hey @michelleyhbn @LucSGeorges just wanted to share that there seem to be errors when using 4x A100 80gbs to host Bixtral on HF endpoints… seems like more than enough RAM, but still crashed — tried gptq etc. No dice, all failed. Not a huge deal for me, but wanted to make sure
-
Cerebras Selected to Forbes AI 50 List for AI Chip
By
–
Cerebras is thrilled to be selected on the 2024 Forbes AI 50! Here are a few reasons why we made the cut: Cerebras is the only AI chip startup on this year’s list. Learn more about our latest generation of hardware, the CS-3: https://
cerebras.net/product-system/ We enable top -
Microsoft Optimizes Bing Ads with NVIDIA Triton Inference
By
–
DYK – By using Triton Inference Server, Microsoft Bing personalized ads to users applying sophisticated techniques to do more work in less time with less computer memory. Read their story to learn more > https://
nvda.ws/3UbmeBw #AI #inference #NVIDIATriton By using Triton -
Incumbent Tech Giants Struggle with Pivoting and Innovation
By
–
“What happens to the big incumbents is that they are handcuffed to their past investments, their financial situation, their staffing, their customer roadmaps,” said @tenstorrent COO @kewitek . “They can’t pivot, and they can’t change.” @ojoyoshida https://
ojoyoshidareport.com/the-audacity-o
f-tenstorrent/
… -
T4 Models Upgraded to 16 GB RAM at Same Price
By
–
T4 models can now use up to 16 GB of RAM (for the same price).
-
Unlock Embedded Applications with Altera FPGA Technology
By
–
Unlock Embedded Applications with Altera FPGA
— Ronald van Loon (@Ronald_vanLoon) 11 avril 2024
by @Ronald_vanLoon & Ben Lee |
Get started now: https://t.co/8dTCzGJkwg#AlteraAmbassador @AlteraFPGA_ #WeAreAltera #AcceleratingInnovators #LeapIntoTheFuture #FPGA #Agilex #AlteraAtEmbeddedWorld #EW24 #Embeddedworld
Cc:… pic.twitter.com/Z2qYAKFv6TUnlock Embedded Applications with Altera FPGA
by @Ronald_vanLoon & Ben Lee | Get started now: https://
bit.ly/43Qz5wd #AlteraAmbassador @AlteraFPGA_ #WeAreAltera #AcceleratingInnovators #LeapIntoTheFuture #FPGA #Agilex #AlteraAtEmbeddedWorld #EW24 #Embeddedworld Cc: -
Edge AI Inference Energy Efficiency Through At-Memory Architecture
By
–
Scaling #AI to the edge sustainably is a challenge. Our CEO Chris Walker joins @MoorInsStrat podcast to explain how our unique at-memory architecture radically cuts data movement energy consumption for edge AI inference. Listen now: https://
x.com/i/broadcasts/1
YqGoDaakVQJv
…