搭载 AMD Gorgon Halo 系统级芯片平台的首批系统已面世，板载高达 192 GB 统一内存。这足以让你在桌面上运行 DeepSeek V4 Flash —— 前提是你愿意为海量内存支付高昂溢价。GMKtec EVO-X5 Pro 是首批搭载“禅宗之家”顶配 Ryzen AI Max+ 495 SoC 的产品之一 —— 你会明白为何我们今后就直接称其为 Gorgon Halo —— 但其常规起售价高达 6799 美元，如此大容量内存绝非廉价之选。
The first systems powered by AMD's Gorgon Halo system-on-chip (SoC) platform have arrived, boasting up to 192 GB of unified memory on board. That's enough to put DeepSeek V4 Flash on your desk if you can afford to pay a hefty premium for all that RAM. GMKtec's EVO-X5 Pro is among the first to feature the House of Zen's top-specced Ryzen AI Max+ 495 SoC — you can see why we're just going to call it Gorgon Halo from here on out — but with regular pricing starting at $6,799, all that memory doesn't come cheap.
不过对于本地 AI 爱好者，或许还有注重隐私的中小企业而言，在家庭和办公室的安全环境中运行更大、更强的模型，这笔投入或许物有所值。在 4-bit 精度下，Gorgon Halo 配备的内存足以在 Linux 上运行约 3450 亿参数模型，或在 Windows 上运行 3200 亿参数模型。差异源于内存分区机制：Windows 需预留 160 GB 显存给 GPU，而 Linux 内核与 AMD GPU 驱动允许用户利用几乎所有可用内存。
But for local AI enthusiasts and perhaps privacy-conscious small businesses, it might be worth it just to run larger, more capable models from the security of their homes and offices. At 4-bit precision, Gorgon Halo is equipped with enough memory to run models to around 345 billion parameters on Linux or 320 billion parameters on Windows. The difference here comes down to memory partitioning.
这让 Gorgon Halo 用户得以触达 2840 亿参数的 DeepSeek V4 Flash、Z.AI 3200 亿参数的 GLM-5.3-Flash 等高知名度前沿模型。
On Windows, you'll need to allocate 160 GB of memory to the GPU, while the Linux kernel and AMD's GPU drivers enable users to take advantage of nearly all of the available memory.
该芯片于今年早些时候发布，Gorgon Halo 本质上是为 AMD AI Halo 工作站提供动力的 Strix Halo APU 的工厂超频版 —— 我们曾在今夏对其进行评测。两款芯片均可配置高达 16 个 Zen 5 核心、40 个计算单元的集成 GPU，以及基于 XDNA 2 的 NPU，为微软强行推送的所有 Copilot+ 功能提供算力。两者的主要区别在于：AMD 将全域频率提升了约 100 MHz，且 Gorgon Halo 支持 192 GB 更快的 8533 MT/s LPDDR5x 内存，而 Strix Halo 仅支持 128 GB 8000 MT/s 内存。就性能而言，我们预期频率提升不会带来显著差异。
This puts high-profile frontier models like the 284 billion parameter DeepSeek V4 Flash or Z.AI's 320 billion parameter GLM-5.3-Flash within reach of Gorgon Halo customers. The chip Announced earlier this year, Gorgon Halo is essentially a factory overclocked version of the Strix Halo APU that powers AMD's AI Halo workstation that we reviewed this summer. Both chips can be had with up to 16 Zen 5 cores, a 40-compute-unit integrated GPU, and an XDNA 2-based NPU to power all the Copilot+ features you didn't ask for but Microsoft is pushing on you anyway. The big difference between the chips is that AMD has bumped up the clocks by about 100 MHz across the board, and Gorgon Halo now supports 192 GB of faster 8,533 MT/s LPDDR5x memory compared to Strix Halo's 128 GB of 8,000 MT/s.
话虽如此，更快的内存对本地 AI 爱好者来说或许能带来一点帮助。大语言模型（LLM）推理——即运行预训练模型的过程——主要受限于内存。这意味着内存越快，系统生成 token 的速度就越快。但虽然 Gorgon Halo 实现了 273 GB/s 的高内存带宽，这仅比 Strix Halo 的 256 GB/s 快约 6.5%。充其量，在高度优化的模型上，你每秒只能多看到几个 token。
In terms of performance, we don't expect the clock bump to make a meaningful difference. With that said, the faster memory could help a little bit for local AI enthusiasts. Large language model (LLM) inference — that is, the process of running pre-trained models — is predominantly memory bound. That means the faster your memory is, the faster the system can generate tokens. But while Gorgon Halo does achieve high memory bandwidth at 273 GB/s, that's only about 6.5 percent faster than Strix Halo at 256 GB/s.
被挤出市场Gorgon 的主要吸引力其实在于更高内存容量的前景。但正如我们之前提到的，持续的内存短缺对消费电子产品并不友好。一年前，一台 128 GB 的 Strix Halo 设备售价在 2000 到 3000 美元之间，具体取决于 OEM 厂商。自那时起 DDR5 价格飙升，等到 AMD 今夏推出其 AI Halo 工作站时，价格已跃升至 4000 美元。
At most, you can expect to see a few more tokens a second on highly optimized models. Priced out of the market Gorgon's main attraction is really the prospect of higher memory capacity. But as we mentioned before, the ongoing memory shortage has not been kind to consumer electronics. A year ago, a 128 GB Strix Halo box would have set you back between $2,000 and $3,000 depending on OEM.
内存多出 50%，基于 Gorgon Halo 的产品价格势必更高。GMKtec 这款产品已贵出 64%，而该品牌通常主打亲民路线。Framework 等更高端品牌，以及惠普 ZBook Ultra G3a 等笔记本，恐将溢价更高。
DDR5 prices have skyrocketed since then and by the time AMD launched its AI Halo workstation this summer, prices had jumped to $4,000. With 50 percent more memory, you can expect Gorgon Halo-based products to be even pricier. The GMKtec unit is already 64 percent more expensive and the brand is usually on the more affordable side of things.
对 AMD 而言不幸的是，这方面几乎无计可施。得益于 AI 热潮，LPDDR5x 供应短缺。英伟达售出的每个 NVL72 机柜就装有 36 到 54 TB 的该类内存，这还只是 CPU 内存。随着 AMD 为其面向 AI 优化的 Verano Epyc 处理器采用 LPDDR5x，局势只会恶化。如果你现在需要大量内存，恐将备受煎熬，且短期内不指望好转。今年早些时候，三星警告称内存供应恐将持续紧张至 2028 年，这对包括英伟达在内的所有人都是坏消息。
More premium brands like Framework and notebooks like HP's ZBook Ultra G3a will likely command a steep premium. Unfortunately for AMD, there's not a lot that can be done. LPDDR5x is in short supply thanks to the AI boom. Each NVL72 rack Nvidia sells is packed with between 36 and 54 TB of the stuff, and that's just CPU memory. The situation is only going to get worse as AMD embraces LPDDR5x for its own AI-optimized Verano Epycs. If you need a lot of memory right now, you're in for a lot of pain, and the situation isn't expected to improve anytime soon.
英伟达基于其 GB10 SoC 的 AI 工作站价格在过去一年里也大幅上涨，从一年前的 3000 美元至 4000 美元跳涨至如今的 6000 美元至 8000 美元，这对其 Windows SoC 首秀并不利好。在今春晚些时候的 Computex 上宣布，英伟达的 RTX Spark 是一系列来自主要 OEM 厂商的高端 Windows 系统，基于同款 GB10 SoC，提供高达 128 GB 的内存容量。
Earlier this year, Samsung warned that memory supplies are likely to remain constrained through 2028, which is bad news for everyone including AMD's top competitor Nvidia. Nvidia has also seen the price of AI workstations based on its GB10 SoC roughly double over the past year, jumping from $3,000-$4,000 a year ago to $6,000-$8,000 today, which doesn't bode well for its Windows SoC debut.
考虑到这些系统将以迷你 PC 和笔记本电脑两种形式提供，我们不会惊讶看到它们比现有的 GB10 机型溢价更高。这再次让 AMD 处于可以通过提供更多内存且价格更低来削弱竞争对手的位置，但当你买不起这两种选择中的任何一种时，货比三家就变得异常困难。®
Announced at Computex late this spring, Nvidia's RTX Spark is a family of premium Windows systems from major OEMs based around the same GB10 SoC and offering up to 128 GB of memory capacity. Considering these systems will be offered as both mini PCs and notebooks, we wouldn't be surprised to see them command a premium over existing GB10 boxes. Once again, this puts AMD in a position to undercut its competitor by offering more memory for less, but it's awfully hard to cross shop when you can't afford either option. ®