#Amd

AMD's Taalas Deal Bets Inference Is the New Bitcoin Mining

Hardwiring weights into silicon only pays when a workload stops moving. AMD is betting inference just did.

Article · 1w ago0

AMD Buys Taalas, Betting Inference Ends Etched in Silicon

Hardwired weights promise 17,000 tokens a second — if model churn doesn't outrun the mask set.

Article · 1w ago1

DeepSeek V4 Flash on One AMD GPU Took Nine Patches

A single 192 GB MI300X now serves the full checkpoint — proof of AMD's hardware edge and its lingering software gap.

Article · 1w ago1

RADV Already Ate Linux. Now Valve Points It at Windows

Collabora's experimental Win32 port runs Counter-Strike 2 atop a reverse-engineered, undocumented AMD kernel interface.

Article · 1w ago0

KNOD Wants to Turn Your AMD GPU Into a DPU

New RFC patches let the Linux kernel JIT-compile packet programs straight onto GPU silicon, with no ROCm required.

Article · 2w ago0

PyTorch Monarch Escapes the CUDA Bubble

AMD and Meta port the single-controller runtime to ROCm, and the fault-tolerance demo is the real story.

Article · 2w ago0

Framework's 192GB Desktop Is a Bet That Local AI Stays Sparse

Fifty percent more memory and seven percent more bandwidth only add up if you run MoE models.

Article · 3w ago2

x86 ACE Specs Matrix AI, Silicon Still Far Off

A joint AMD-Intel outer-product extension freezes the ISA contract, but real cores are years away.

Article · 4w ago1

VRAM Beats TOPS for 2026 Local AI GPUs

Memory capacity and bandwidth decide what models you can run; software maturity still decides how painful it feels.

Article · 1mo ago2

AMD's $4,000 AI Halo: Breaking the VRAM Wall at a Premium

The Ryzen AI Halo dev kit offers 128GB of unified memory for local LLMs, but the price tag hurts.

Article · 1mo ago2

AMD's GLM-5.2 win over Blackwell is a software story

MI355X hits about 80% of B200 throughput at over 2x lower cost, but only after hand-tuning the ROCm stack.

Article · 1mo ago5

ZLUDA 6 Arrives: What a Return to Hobbyist Status Means for CUDA Translation

The translation layer can run unmodified CUDA binaries on AMD and Intel GPUs, but its new direction changes the enterprise calculus.

Release · 1mo ago5

Disaggregating LLM Inference: Inside AMD's ATOM and ATOMesh Stack

AMD's native ROCm serving stack splits prefill and decode to eliminate head-of-line blocking on Instinct hardware.

Article · 1mo ago0

AMD's TSME Reversal Is a Lesson in Firmware Governance

Restoring memory encryption on consumer Ryzen 9000 fixes the symptom, but the silent BIOS-level segmentation is the real problem.

Article · 1mo ago5

AMD Strips TSME Memory Encryption From Consumer CPUs

Developers relying on consumer Ryzen hardware for secure edge or workstation deployments must re-evaluate their physical trust boundaries.

News · 1mo ago2