Nvidia’s new CUDA Tile sparks chatter about the end of software lock-in
Tiling makeover could open the door to rival GPUs
Nvidia has rolled out one of the biggest updates to its CUDA software stack in years, and chip design legend Jim Keller reckons it might even spell the end of its long-guarded exclusivity.
AMD bangs open source drum to prise devs away from Cuda
Pitches ROCm as the antidote to Nvidia’s walled garden
AMD reckons it can win over AI developers by offering an open software ecosystem rather than locking everything behind a proprietary moat.
Cerebras scoops $1bn to take on Nvidia
Silicon Valley start-up wants to crash the AI chip party
Chipmaker Cerebras Systems has bagged more than $1 billion from backers including Fidelity and Donald Trump Jr’s 1789 Capital as it tries to peel customers away from Nvidia.
AMD gets PyTorch working on Windows
Radeon 9000, 7000 and Ryzen AI kit now less useless
AMD has delivered on its Computex 2025 promise to make PyTorch work on Windows for consumer GPUs and APUs.
AMD waves ROCm 7 about to take on Nvidia’s CUDA cult
Hopes devs finally look at its software stack
AMD has shoved ROCm 7 onto GitHub, fuelling speculation that this might be the release to finally lead developers out of Nvidia’s CUDA walled garden and into a promised land.
Nvidia makes CUDA play nice with RISC-V
GPU giant eyes open silicon for edge
At the 2025 RISC-V Summit in China, Nvidia dropped a bombshell by revealing that its CUDA software platform will now run on RISC-V CPUs.
AMD snaps up Brium to loosen Nvidia’s AI iron grip
Chipmaker adds stealthy startup to its growing anti-Nvidia arsenal
AMD has snapped up a little-known AI startup called Brium in a move that looks suspiciously like an attempt to smash Nvidia’s stranglehold on the AI software ecosystem.
Nvidia expects to have used a billion RISC-V cores this year
It is not all about CUDA
By the end of the year Nvidia will have shipped a billion RISC-V cores under the bonnet of its GPUS.
Nvidia’s CUDA faces a worthy foe
AMD’s stealthy gambit
AMD has made a significant move in the world of GPU software development by introducing a solution that bridges the gap between its ROCm technology and NVIDIA’s CUDA binaries.
Lamini ditches Nvidia
AMD GPUs easier for Llama-2
AI startup Lamini uses AMD GPUs instead of the more popular Nvidia GPUs to run large language models (LLMs) like Llama-2 for customers.