Recent Posts#
vLLM-ATOM: Unlocking Native AMD Performance in the vLLM Ecosystem
Use ATOM as an out-of-tree vLLM plugin to keep vLLM compatibility while enabling AMD-optimized attention, model execution, and multi-model support including Kimi-K2.5.
AMD-Powered 3D Gaussian Splatting for Autonomous Driving Scenes
Run Street Gaussians on AMD Instinct MI300: migrate to latest gsplat, install on ROCm, and render dynamic street scenes.
Accelerating Mixture-of-Experts Execution with FarSkip-Collective Models
Explore a new MoE architecture designed for native computation-communication overlap, enabling efficient distributed execution.
TraceLens: Democratizing AI Performance Analysis
Explore how TraceLens automates profiler trace analysis to pinpoint bottlenecks and optimize AI workloads.
Primus Projection: Estimate Memory and Performance Before You Train
Learn how to use the Primus projection tool to estimate memory and performance for large-scale LLM training on AMD Instinct™ accelerator platforms.
Styled Text Image Generation with Eruku on AMD
Hands-on, reproducible guide to train and run Eruku on LUMI supercomputer, powered by AMD Instinct MI250X GPUs.
Getting Started with FlyDSL Nightly Wheels on ROCm
A practical guide to installing and using FlyDSL nightly wheels on ROCm for fast, Python-native GPU kernel development
FLy: A New Paradigm for Speculative Decoding — Accepting Semantically Correct Drafts Beyond Exact Match
This blog explores a new training-free loosely speculative decoding method, that can accept mismatches that are semantically valid and speedup original SPD method.
Serving CTR Recommendation Models with Triton Inference Server using the ONNX Runtime Backend
Learn how to deploy AI models on AMD GPUs with Triton Inference Server, now supporting ONNX Runtime and Python backends, and see performance benchmarks.
FlashInfer on ROCm: High‑Throughput Prefill Attention via AITER
FlashInfer is an open-source library for accelerating LLM serving that is now supported by ROCm.
Customizing Kernels with hipBLASLt TensileLite GEMM Tuning - Advanced User Guide
Master hipBLASLt TensileLite Tuning. Learn to build custom kernels that deliver 150%-250% faster GEMM performance on AMD Instinct™ MI300X GPUs