ROCm Blogs
Technical blog from AMD with in-depth articles on ROCm, GPU-accelerated AI, and HPC
| What is it | Technical blog from AMD with in-depth articles on ROCm, GPU-accelerated AI, and HPC |
|---|---|
| Pricing | Unknown |
| Platform | Web Application |
| API | Yes |
| Best for | learning GPU programming and optimization on AMD hardware, optimizing AI model performance on Instinct GPUs |
| Domain registered | 1986 |
Data updated Aug. 1, 2026
What does ROCm Blogs do?
ROCm Blogs is AMD's official technical blog for the ROCm platform. It publishes regular articles covering GPU-accelerated computing, artificial intelligence, machine learning, and high-performance computing. The content ranges from deep dives into MLPerf benchmark submissions to step-by-step guides for running large language models on Radeon and Instinct GPUs. Each post is written by AMD engineers and includes code snippets, performance data, and optimization techniques.
The blog is organized into clear sections: Featured Posts, Recent Posts, and curated categories like Ecosystems & Partners, Applications & Models, and Software Tools & Optimizations. Articles often include reproducible instructions and links to source code, making it practical for developers who want to test the techniques themselves. Recent topics include speculative decoding with EAGLE3, MXFP6 quantization, and building custom hipBLASLt libraries. The blog also covers partnerships with frameworks like JAX, PyTorch, and Ray.
This blog is most useful for developers, researchers, and AI practitioners working with AMD GPUs. It provides the latest information on ROCm releases, performance optimization tips, and real-world deployment patterns. Whether you are tuning an LLM inference pipeline, porting a workload to ROCm, or exploring quantum computing with Qiskit Aer, the articles offer concrete, actionable guidance.
Key features
What makes it stand outWho is ROCm Blogs for?
Who benefits most from this toolTrust & presence
Similar tools
vLLM
Verified
AI inference
High-throughput LLM inference engine for fast, memory-efficient AI model serving.
RunInfra
Verified
AI inference
Optimize open-source AI models for production — benchmark engines, tune latency, and deploy on any GPU.
RightNow AI
Verified
Developer Tools
AI-powered code editor for GPU kernel development, optimization, and real-time profiling.
Groq
Verified
AI inference
High-speed, low-cost AI inference API for running large language models with minimal latency.
LLMTest
Verified
Testing
Automatically optimizes prompts and AI models in your app for better performance and lower costs.
General Compute
Verified
AI inference
High-speed AI inference API powered by purpose-built hardware, not repurposed GPUs.
SelfHostLLM
Verified
LLM
Calculate GPU memory and performance requirements for running large language models on your own hardware
RunLLM
Verified
Developer Tools
Aqueduct automates the engineering required to take data science to production. By abstracting away low-level cloud infrastructure, Aqueduct enables data teams to run models anywhere, publish predictions where they're needed, and monitor results reliably.