amd

Analyzes the common characteristics of the best-performing tech stocks over the past decade (e.g., Nvidia, AMD), including deep ties to major tech waves, hardware-to-recurring-revenue transitions, significant drawdowns, star CEOs, and points out that embodied intelligence and robotics are the most promising directions in AI, where ordinary people should invest early.

Wafer successfully runs GLM-5.2 on AMD MI355X with competitive performance, challenging NVIDIA's CUDA moat as AI agents automate software optimization, potentially eroding NVIDIA's high premium.

Wafer demonstrates that AMD MI355X GPUs offer competitive inference performance for frontier models like GLM5.2 at significantly lower cost than NVIDIA Blackwell, achieving 80% of B200 throughput at under half the price, using MXFP4 quantization and sglang.

Engineers successfully serve GLM 5.2 on AMD MI355X at 2626 tok/s per node and 213 tok/s single stream, achieving ~80% of B200 throughput at over 2x lower cost than Blackwell.

This article explains that for RAG applications, prefill speed matters more than decode speed, and discusses why AMD's Strix Halo APU struggles with interactive use cases.

AMD releases Micro-World, an action-controlled interactive world model built on the Wan2.1 family, with open-source weights, code, and a curated dataset for controllable world modeling.

A multi-agent synthetic data pipeline with SFT and GRPO RL post-training improves HIP compilation and correctness on AMD MI350X GPUs for a 14B open-source model.

A setup guide for using a custom Docker/Podman toolbox with ROCm/RCCL RDMA support to cluster two AMD Strix Halo nodes, enabling vLLM with tensor parallelism across 256GB unified memory.

Discusses the potential value of using two Radeon RX 9060 XT 16GB graphics cards in a CrossFire configuration, likely weighing performance vs. cost.

An AMD engineer contributed an ONNX Runtime backend to FFmpeg's DNN filter, enabling AI model inference on GPU and NPU platforms for tasks like upscaling and object detection, notably making Ryzen AI NPU useful for FFmpeg.

The NPU on AMD Strix Halo devices is now usable for AI inference, enabling hybrid mode that combines NPU and iGPU for faster prompt processing. Tools like Lemonade and AMD's ROCm software make this possible.

After user outcry over the silent removal of TSME memory encryption from consumer Ryzen processors, AMD has announced it will reinstate the feature in a July firmware update.

Valve is working with AMD to bring FSR 4 upscaling to the Steam Machine, promising significant graphical improvements for the console.

Technical report on running Qwen 3.6 27B Q8 model on a dual AMD Radeon R9700 setup using llama.cpp with ROCm, including performance benchmarks and configuration details.

A comparison of AI inference frameworks ROCm, Vulkan, and vLLM running on dual AMD Radeon 9700 GPUs, likely benchmarking performance for large language models.

A detailed comparison of local AI hardware in terms of memory capacity, bandwidth, and software stack, covering GPUs, Apple Silicon, AMD, Intel, Tenstorrent, and others, with a focus on what bottlenecks matter for AI inference.

AMD will reinstate Transparent Secure Memory Encryption (TSME) on non-PRO Ryzen 9000 desktop CPUs via a BIOS update in July, following community feedback after the feature was silently removed in an earlier firmware update.

The article details a setup running six AI agents 24/7 on a Minisforum MS-S1 Max mini workstation with AMD Ryzen AI Max+ 395 chip, costing $11/month in electricity. It highlights the shift from cloud API costs to local inference, enabling always-on agents for tasks like email sorting, research monitoring, and document processing.

Discussion about upcoming AMD GPU offerings and their potential for building an LLM rig, asking the community for build suggestions.

AMD has released GAIA 0.21.2, introducing gaia-bash, an AI-powered bash scripting assistant for writing, reviewing, testing, and debugging shell scripts on AMD hardware. It supports multiple interfaces including TUI, CLI, pipe mode, REST API, and MCP stdio server.