Instinct AI accelerators and CPUs
Headquartered in Santa Clara, California, AMD makes CPUs, GPUs and the Instinct line of data-centre AI accelerators with its ROCm software stack, and is NVIDIA's main rival in AI chips.
AMD MI350 support matters because GPU kernel tooling has historically centred on NVIDIA's stack, and this gives teams serving open models a Python-native path to tune kernels without CUDA-only tools. Benchmark and deployment engineers running vLLM on AMD accelerators can adopt the DSL directly from PyTorch's ecosystem and measure the 3.01x serving gain against their own baselines.
AMD 和英特尔这对老对手 10 月 20 日宣布联手，以后跑 PyTorch 不用再分 AMD 还是 Intel CPU 单独调优了，搞部署的可以看看。
云算力公司 Modal 第一次用上英伟达的 Vera CPU，测试显示智能体吞吐量翻倍，CPU 市场有新玩家了。
威联通一次发了三件套：8 盘位和 12 盘位的 USB4 JBOD 扩展柜，再加一张双口 40Gbps 的 PCIe 扩展卡，NAS 存储不够用的可以看看。
AMD 把剪枝当成多目标优化来做，Llama-3.1-70B 砍掉 25% 层后数学分还能涨 10 个点，代码也会开源。
RX 6000 老卡用户可以白嫖 FSR 4.1 超分了，装这个驱动就能用，不过《赛博朋克 2077》开帧生成会有崩溃，注意看已知问题列表。
做 LLM 部署的可以看看，EAGLE-4 开源了，推测解码能自适应硬件和负载，Qwen3.8-27B 实测跑到 961.6 tokens/s。
vLLM now spans three hardware ecosystems, which reduces the lock-in that previously tied inference stacks to NVIDIA GPUs. Teams running self-hosted inference can benchmark vLLM against vendor-specific runtimes on AMD and TPU infrastructure. The KV cache management work matters most for serving long-context models where memory, not compute, is the bottleneck.
AMD 把 FSR 4.1 带到 Xbox 主机和掌机上了，RX 6000 显卡今天就能用，老设备画面也能跟着升级。
The shift from AI PC to Agentic PC changes the buying criteria themselves: local model reliability and agent execution replace CPU and GPU specs as the deciding factors. Vendors in China's supply chain, including Lenovo and Huawei, will need to prove task-completion capability, which shapes what lands in ASEAN retail channels next. Procurement teams comparing laptop lines should start asking for local-model benchmarks instead of only hardware sheets.
Steam Deck 2 的芯片疑似现身跑分库：6 核 12 线程加 24GB LPDDR5，内存速率从 6400MT/s 拉到 9600MT/s，掌机党可以看看具体提升多少。
The same serving stack now optimises both NVIDIA and AMD paths, giving operators a concrete reason to benchmark MI355X deployments against GB300 rather than defaulting to CUDA. Teams running agentic inference traces can enable the mode through SGLang and expect P90 latency gains without changing models. The plugin system also means third-party kernels — starting with GLM-5.3, K3, and MiniMax M3 — will arrive without waiting for a core release.
