tldrdailynews.com logo
Tag

Rocm

All articles tagged with #rocm

AMD Unveils Helios Rack and Open AI Strategy at Advancing AI 2026
technology2 hours ago

AMD Unveils Helios Rack and Open AI Strategy at Advancing AI 2026

AMD’s Advancing AI 2026 keynote previews Helios, a rack-scale AI system uniting MI455X GPUs, Venice CPUs, and Pensando DPUs, with production shipments starting in late 2026 and ramping in 2027; ROCm.AI and an open software stack are pushed alongside partnerships with Anthropic, OpenAI, Meta, Cerebras, Cisco and AT&T to scale agentic AI across data centers. The company projects a $1.4 trillion AI accelerator TAM by 2030 and outlines aggressive CPU/GPU roadmaps (Venice HF, Verano, Zen 6, MI500), along with physical AI devices, emphasizing co-design with customers and an open ecosystem.

via ServeTheHome|
#ai#amd#datacenter
technology17 days ago

AMD's Debian-Powered Ryzen AI Developer Platform Targets Linux AI Development

AMD's Ryzen AI Halo developer PC ships with a Debian-based OS named the Ryzen AI Developer Platform 1 "Rex", featuring a GNOME desktop and a GUI "Ryzen AI Developer Center" to manage ROCm, Llama.cpp, vLLM, PyTorch and other AI tools, plus telemetry controls and remote SSH; it includes ROCm 7.13 preview and AMD apps, delivering a polished Linux AI development environment atop Debian rather than Ubuntu.

via Phoronix|
#ai-development#amd#debian
AMD Bets on the AI Edge with Ryzen AI Halo Mini PC Against DGX Spark
technology1 month ago

AMD Bets on the AI Edge with Ryzen AI Halo Mini PC Against DGX Spark

AMD launched the Ryzen AI Halo Mini PC at $3999, packing a Ryzen AI MAX+ 395 SoC (Zen 5, 16 cores, 50 TOPS NPU), 128 GB RAM, 2 TB storage in a compact 5.9"x5.9"x1.7" chassis with ROCm 7.2.2 and a Dev-ready software stack. It positions itself as a cheaper, capable alternative to NVIDIA’s DGX Spark (around $4679), claiming higher token throughput and strong local AI economics (potentially breaking even in ~6 months vs cloud costs) while teasing a future MAX+ 495 with 192 GB RAM for even larger models in Q3 2026. Availability starts in June.

via Wccftech|
#ai-pc#amd#dgx-spark
AMD Poised for a 2026 Upside as It Gains Ground on Nvidia in AI Infrastructure
business5 months ago

AMD Poised for a 2026 Upside as It Gains Ground on Nvidia in AI Infrastructure

AMD is gaining traction in the data-center AI race by supplying hyperscalers with its Instinct accelerators and open-source ROCm software, offering a flexible alternative to Nvidia’s CUDA. With major buyers like Microsoft, Meta, Oracle, and OpenAI deploying AMD alongside Nvidia, the company could see stronger deal flow, improved unit economics, and margin expansion as AI infrastructure spend—driven by hyperscalers—approaches hundreds of billions of dollars, potentially making 2026 a transformative year for AMD.

via The Motley Fool|
#ai-chips#amd#business
technology2 years ago

AMD's AI-Powered Ryzen Chips Set to Revolutionize PC Performance

AMD is set to release a Unified AI Software Stack by the end of the year, aiming to optimize AI workloads across CPUs, GPUs, and NPUs. This stack will utilize MLIR for intermediate representation and is expected to enhance AI workload management and developer experience. While details are still emerging, the stack promises to integrate with existing AMD software like ROCm and could potentially support non-AMD products.

via Phoronix|
#ai#amd#mlir
"AMD GPUs Now Support Unmodified NVIDIA CUDA Apps with ROCm and ZLUDA"
technology2 years ago

"AMD GPUs Now Support Unmodified NVIDIA CUDA Apps with ROCm and ZLUDA"

AMD's ROCm now indirectly supports NVIDIA CUDA libraries through the open-source ZLUDA project, allowing AMD GPU owners to run CUDA apps within ROCm without code adaptation. ZLUDA, originally designed for NVIDIA CUDA support on Intel GPUs, has been successfully tested with AMD's Radeon GPUs on ROCm, enabling performance on par with NVIDIA for rendering tasks. While native CUDA support on Radeon GPUs remains unavailable, the potential of ZLUDA in bridging ROCm and CUDA stacks for AI applications is promising, with the developer exploring NVIDIA's upscaling capabilities on RDNA GPUs.

via Wccftech|
#amd#cuda#nvidia
technology2 years ago

"AMD's Open-Source CUDA Implementation Funded Quietly"

AMD quietly funded a project to create a drop-in replacement for CUDA on AMD GPUs, allowing many CUDA applications to run on the ROCm stack without modification. The project, based on the ZLUDA implementation, was developed by Andrzej Janik and is now open-source, dual-licensed under Apache 2.0 or MIT. While it's a promising solution, it's not without limitations, such as incomplete NVIDIA OptiX support. The implementation, leveraging the Rust programming language, has been tested successfully, and the exposed device name for Radeon GPUs via CUDA will be updated as part of the open-sourcing.

via Phoronix|
#amd#cuda#open-source
technology2 years ago

AMD's Open-Source Advancements and AI Accelerator Launches Set for December Event

AMD is teasing open standards and open-source ahead of its "Advancing AI" event scheduled for December 6th. While details about the event are scarce, speculation suggests that it could involve the rollout of ROCm for more GPUs or the introduction of open-source/Linux support for Ryzen AI. AMD's promotion of open elements has sparked interest among enthusiasts, who are eager to see what announcements the event will bring.

via Phoronix|
#advancing-ai#amd#open-source