Intel unveils the Arc B390 integrated GPU, offering improved gaming performance, and says it will power gaming handhelds built by MSI, Acer, Microsoft, and more
AMD unveils the 12-core Ryzen AI Max+ 392, and the 8-core AI Max+ 388 processors, both with 40 graphics compute units and offering 60 TFLOPS of GPU performance
AMD unveils Ryzen AI 400 Series chips for AI PCs, with up to 12 Zen 5 CPU cores and 16 RDNA 3.5 GPU cores, built with TSMC's N4X node, available in Q1 2026
Lemon Slice, which launched a 20B-parameter model that creates digital avatars for AI agents, streaming video at 20 FPS on a single GPU, raised a $10.5M seed
Developers and companies are increasingly deploying AI agents and chatbots within their apps, but so far they've mostly been restricted to text.
A look at Google's TPU evolution, its decision to sell chips to competitors after 12 years of internal use, the TPU vs. GPU architecture showdown, and more
Is GPU vs TPU the CISC vs RISC contest of the modern era? … All of a sudden, everyone is talking - and writing - about Google's TPUs (Tensor Processing Units). LinkedIn: Aditya Harit LinkedIn: Aditya ...
Samsung unveils the Exynos 2600, the world's first smartphone SoC built on a 2nm Gate-All-Around process, expected to power some Galaxy S26 and S26 Plus models
Get ready for a new mobile experience. Will Sattelberg / 9to5Google : Galaxy Z Flip 8 could remain exclusive to Samsung's Exynos 2600, report suggests The Indian Express : Samsung introduces Exynos 26...
Mira Murati's Thinking Machines Lab makes Tinker, its API for fine-tuning language models, generally available, adds support for Kimi K2 Thinking, and more
Tinker is a dream for multi-agent setups, Nathan Lambert / @natolambert : Please add olmo3 @johnschulman2 et al. The goal is to make it the foundational research infrastructure for academic LLM work. ...
DeepSeek releases DeepSeek-OCR, a vision language model designed for efficient vision-text compression, enabling longer contexts with less compute
the new frontier of OCR from @deepseek_ai , exploring optical context compression for LLMs, is running blazingly fast on vLLM ⚡ (~2500 tokens/s on A100-40G) — powered by vllm==0.8.5 for day-0 model su...
Intel unveils Crescent Island, a data center GPU “power- and cost-optimized” for AI inference, featuring its Xe3P microarchitecture and 160GB of LPDDR5X memory
Dylan Martin / CRN :
Apple debuts its M5 chip, with a 10-core GPU, a Neural Accelerator in each core, offering 4x+ the performance of M4, and a 10-core CPU with six efficiency cores
M5 delivers over 4x the peak GPU compute performance for AI compared to M4, featuring a next-generation GPU with a Neural Accelerator …