Type: semiconductor company
AMD is an American semiconductor company that designs CPUs and GPUs, including the Instinct accelerators competing in the AI training and inference hardware market. It is a key alternative in the AI compute supply chain. GROUNDING tracks AMD’s AI accelerators, software stack, and hardware roadmap.
Recent Updates
- 2026-06-28: Guide for a two-node AMD Strix Halo RDMA cluster running vLLM (Hacker News) · github.com — Intel · Fedora · Framework · vLLM · Ray · ROCm
- 2026-06-29: Budget local AI builds under 100k rubles are benchmarked for CPU and cheap GPU inference (Все статьи подряд / Искусственный интеллект / Хабр) · habr.com — Long Context LLM Evals MSI ASUS · Tesla Ryzen 7 5800X Threadripper 2920X CMP 40HX Tesla V100
- 2026-06-29: Building a Home AI Server on a Budget (Все статьи подряд / Искусственный интеллект / Хабр) · habr.com — Long Context Habr · Ubuntu KDE · Docker · llama.cpp · vLLM · ROCm · Vulkan · Qwen3.6-27B
- 2026-07-03: Controlled study finds lightweight CNN winners vary by dataset and hardware (cs.LG updates on arXiv.org) · arxiv.org — LLM Evals NVIDIA · PyTorch EfficientNetV2-S RepViT-M1.0 EfficientNet-B0 MobileNetV3-Small MobileNetV4-Conv-S SqueezeNet1.1 CIFAR-10 CIFAR-100 Tiny ImageNet
- 2026-07-03: GLM5.2 serving on AMD MI355X with claimed 2626 tok/s/node (Hacker News) · wafer.ai — Context Engineering LLM Evals Wafer NVIDIA · Z ai TensorWave · Artificial Analysis · vLLM Atom · SGLang Quark · GLM5.2
- 2026-07-04: MSI Center named pipe flaw could grant SYSTEM-level control (Hacker News) · mrbruh.com — MSI ASUS
- 2026-07-06: vLLM: high-performance open-source LLM inference and serving engine (GitHub AI Ranking Changes (Top 10)) · github.com — Open Source LLMs UC Berkeley · Hugging Face · NVIDIA · Google · Intel · IBM · Huawei Rebellions · Apple Metax Woosuk Kwon Zhuohan Li Siyuan Zhuang Ying Sheng Lianmin Zheng Cody Hao Yu Joseph E. Gonzalez Hao Zhang Ion Stoica · LLaMA · Qwen · Gemma · Mixtral · DeepSeek-V3 Qwen MoE · GPT-OSS · Mamba · Qwen3.5 · LLaVA · Qwen-VL Pixtral E5-Mistral · GTE ColBERT Qwen-Math
- 2026-07-10: Unified memory and memory bandwidth explain why mini PCs can load 70B models (Hacker News) · vettedconsumer.com — NVIDIA · Apple · Intel · Qualcomm NotebookCheck TechPowerUp Chips and Cheese Williams Waterman Patterson Pope
- 2026-07-10: Hands-On with the AMD Ryzen AI Halo (Hacker News) · microcenter.com — Long Context Micro Center Dan Ackerman Jacob Bobo GPT-OSS 120B Qwen3-Coder-30B · Gemma 4 31B Z-Image-Turbo
- 2026-07-13: Signed Symmetric Quantization for Few-Bit Integers (cs.LG updates on arXiv.org) · arxiv.org — Qwen3 · Qwen3.5 · Llama3
- 2026-07-13: STEEL brings FlashAttention-style inference to AMD XDNA NPUs (cs.AI updates on arXiv.org) · arxiv.org
- 2026-07-13: A Strix Halo home AI platform running 34 containers and local embedding/rerank workloads (Все статьи подряд / Искусственный интеллект / Хабр) · habr.com — RAG Embeddings Reranking Vector Database Beelink Dify · RAGFlow · Milvus · Qdrant · Weaviate · MySQL · Elasticsearch · MinIO · PostgreSQL · Redis · Prometheus · Grafana Loki Alloy cAdvisor Alertmanager Authelia · Traefik · n8n · Open WebUI Portainer DGX Spark · Docker · llama.cpp · Vulkan RADV · Qwen3.6-35B-A3B bge-m3-Q8_0 bge-reranker-v2-m3-Q8_0
- 2026-07-20: Moonshine streams PC games to Moonlight clients on Linux (Hacker News) · github.com — NVIDIA · Intel · Arch Linux Moonlight
- 2026-07-22: At WAIC, Taichu Yuanqi showed a heterogeneous compute platform built for agent-era workloads (量子位) · qbitai.com — Agents Tool Use 量子位 · QbitAI 太初（杭州）集成电路有限公司 太初元碁 华为 · 沐曦 阿里平头哥 国盛证券 阿里云 河南空港智算中心 瑞莱智慧 安势信息 清源智契 典枢科技 数安信 龙芯 申威 飞腾 百度 · 清华大学 湖南大学 山东大学 东润 百度飞桨 · 上海人工智能实验室 量旋科技 允中 杨晋喆 苏姿丰 徐世真 Megatron-LM DeepSpeed PaddleHelix · PyTorch · vLLM TecoQSim RiverONE AtomWorld · AlphaFold3 CrossDNA Intern-S1
- 2026-07-22: A 4-Mac Studio cluster with 1.5 TB of shared memory for local AI experiments (Все статьи подряд / Искусственный интеллект / Хабр) · habr.com — Apple DeskPi · NVIDIA · DeepSeek-V3 Kimi K2 Thinking
- 2026-07-22: GigaToken claims ~1000x faster language model tokenization (Hacker News) · github.com — Hugging Face · Apple Qwen/Qwen3-8B · LLaMA-3 · Llama 3.1 · Llama 3.2 · Llama 3.3 Qwen 2 · Qwen 2.5 · Qwen 3 · DeepSeek-V3 · DeepSeek-V3.1 · DeepSeek-V3.2 · DeepSeek R1 · DeepSeek-V4-Flash · DeepSeek V4 Pro GLM-4 GLM 4.1V GLM-4.5 · GLM-4.7 · GLM-5 · GLM-5.2 · GLM-4.7-Flash Nemotron 3 · Nemotron 3 Nano · Nemotron 3 Super · Nemotron 3 Ultra · Kimi K2 · kimi-k2.5 · kimi-k2.6 · Kimi K2.7 Phi-4-mini Phi-4-multimodal · TinyLlama · Phi-3
- 2026-07-25: PyTorch Monarch is ported to AMD Instinct GPUs with ROCm (Hacker News) · pytorch.org — PyTorch · ROCm · CUDA RCCL NCCL · Slurm · Kubernetes SkyPilot DeepSeekV3-671B
FAQ
What is AMD?
AMD is an American semiconductor company that designs CPUs and GPUs, including the Instinct accelerators competing in the AI training and inference hardware market. It is a key alternative in the AI compute supply chain. GROUNDING tracks AMD’s AI accelerators, software stack, and hardware roadmap.
What does this page track?
Dated radar mentions, source links, related concepts, and builder-relevant context for AMD, collected automatically by GROUNDING.
When was AMD last mentioned?
AMD was most recently mentioned in a radar update dated 2026-07-25.
Category: AI Chips & Semiconductor Hardware