Search All Blogs
News and Announcements
-
Meet the Next AMD Ryzen™ AI Halo
Meet the next AMD Ryzen™ AI Halo developer platform, powered by the AMD Ryzen™ AI Max+ PRO 495 processor. Run larger local LLMs, up to 300B parameters…
October 12, 2026
-
-
AMD AI NIC™ Technology Driving AI-Ready Ethernet
Learn how AMD AI NIC™ technology helps organizations build AI-ready Ethernet networks with open standards, programmability, and transport innovation.
October 11, 2026
-
MangoBoost Delivers Open AI Models Powered by AMD
Mango Inference a serverless inference service providing production-ready access to open models powered by AMD Instinct™ GPUs and AMD ROCm™ software
October 09, 2026
-
Rapid Validation on AMD Versal Silicon without Leaving MATLAB or Python
AMD Vitis™ Hardware in the Loop (Vitis HIL) bridges simulation and hardware bring-up by enabling system-level verification on evaluation hardware dire…
October 07, 2026
-
How Agentic AI Is Expanding the Role of the DPU
As agentic AI drives more stateful workloads, DPUs are becoming critical for networking, storage, KV-cache management, and infrastructure efficiency.
October 05, 2026
-
Rightsizing Cloud Workloads on AMD Part 1 Foundations
Learn the foundations of rightsizing cloud workloads on AMD: define goals, choose the right time, classify workloads, and select relevant benchmarks.
October 01, 2026
-
AMD Expands Automotive-Grade FPGAs with Spartan UltraScale+ XA FPGAs for ADAS Applications
The Spartan™ UltraScale+™ XA device joins the AMD portfolio of automotive-grade, safe, and highly scalable FPGAs and adaptive SoCs.
September 29, 2026
Technical Articles and Blogs
-
Kimi K3 on AMD Instinct GPUs with TokenSpeed
TokenSpeed delivers fast, agentic LLM inference on AMD Instinct™ MI355X GPUs, with optimized kernels and leading Kimi K3 performance.
October 09, 2026
-
scikit workloads with AOCL Data Analytics
Accelerate supported scikit-learn workloads with AOCL-Data Analytics while keeping your existing Python code and familiar scikit-learn APIs.
October 09, 2026
-
MoRI UMBP Empowers AMD Instinct GPUs
See how UMBP, MoRI and SGLang optimize KV cache reuse to boost AMD Instinct™ MI355X GPU agentic AI performance and cost efficiency
October 09, 2026
-
Pre-Training Hybrid LLMs from Scratch in Primus on AMD Instinct™ GPUs — ROCm Blogs
Pre-train hybrid LLMs mixing MLA attention with Gated DeltaNet, Kimi Delta Attention and Mamba2 mixers on AMD Instinct MI355X GPUs, with open recipes.
October 07, 2026
-
DI Series: Serving Kimi-K3 MXFP4 Wide-EP Disaggregated on AMD Instinct MI300X and MI325X — ROCm Blogs
Learn how to serve a 1.4 TiB Kimi-K3 MXFP4 mixture-of-experts model disaggregated across AMD Instinct MI300X and MI325X GPUs.
October 07, 2026
-
AMD Instinct MI455X vs MI355X: A Technical Look at the Advancing AI 2026 Inference Numbers — ROCm Blogs
Explore the MI455X vs MI355X inference numbers from Advancing AI 2026, and learn how each benchmark was measured and what it really means.
October 07, 2026
-
Completing the GPU Performance Picture: Understanding TAF Alongside Peak FLOPs and MAF — ROCm Blogs
Learn how Typical Attained FLOPs complements Peak FLOPs and Max-Achievable FLOPs, including the methodology and MI325X results
October 05, 2026
-
ROCm 10.1: Breaking the Data-Movement Bottleneck — ROCm Blogs
ROCm 10.1 targets the data-movement bottleneck with AMD Infinity Storage, NUMA-aware memory, and a modernized stack.
October 04, 2026
