cd /news/ai-infrastructure/sipeed-announces-maixstation-local-a… · home › topics › ai-infrastructure › article
[ARTICLE · art-150573] src=sipeed.com ↗ pub= topic=ai-infrastructure verified=true sentiment=· neutral

Sipeed announces MaixStation local AI computer

Sipeed announced MaixStation, a local AI computer it claims beats Nvidia's DGX Spark, offering 256 dense TOPS and 273 GB/s of memory bandwidth in a 363 cm³ enclosure. Sipeed said MaixStation scales from 27B to 552B models, runs text-to-video and image-to-video locally, and connects multiple units with a single cable, whereas DGX Spark requires a 200G/400G switch costing about $4,000 more beyond two units. Sipeed framed the launch as breaking a monopoly, noting DGX Spark's price rose from $3,999 a year ago to $6,999 today and that its 1,000 TOPS figure is sparse FP4 with 500 dense on paper.

read2 min views15 publishedOct 15, 2026
Sipeed announces MaixStation local AI computer
Image: source

What is the best local AI?什么才是最好的本地 AI?

Is DGX Spark the only answer?DGX Spark 是唯一的答案吗?

$3,999 a year ago.一年前 $3,999,$6,999 today.今天 $6,999,Tomorrow?明天呢?

The new 64GB model costs more than the old 128GB.新出的 64GB,

比当初的 128GB 还贵。

"1,000 TOPS." Sparse. On paper."1000 TOPS",

是纸面上的稀疏算力。

And we were told to just accept it.而我们,只能默默接受。

Not anymore.不必再这样了。

We're breaking the monopoly. Win or lose.我们要打破垄断。

无论成败。

Freedom for personal local AI.为了个人本地 AI 的自由。

From 27B to 552B Models.

From 3" Module to Cluster.*Beats DGX Spark.*从 27B 到 552B 大模型。

从 3 寸核心板,到迷你机架集群。超越 DGX Spark。

363 cm³. Small enough for your pocket.363 cm³,小到可以塞进口袋。

1,000 is sparse FP4, which real models can't use. Dense is 500 on paper, and on real models Spark puts only a small slice of it to work. MaixStation turns far more of its 256 into actual speed.1000 是稀疏 FP4 算力,真实模型用不上。稠密算力纸面也只有 500,跑真实模型时 Spark 只用上其中一小部分。MaixStation 能把 256 中大得多的比例变成实际速度。

Bandwidth sets the ceiling. Architecture decides how close you get. Every token reads the whole model once, and MaixStation keeps more of those 273 GB/s busy.带宽决定上限,架构决定离上限有多近。每生成一个 token 都要把模型读一遍,MaixStation 把这 273 GB/s 用得更满。

Those runs use MTP or speculative decoding. This chart turns MTP off on both sides for a like‑for‑like baseline. Turn it on, and MaixStation gets a similar speed‑up.那些数字开了 MTP 或投机解码。本图两边都关闭 MTP,用标准速率对比。开启 MTP 后,MaixStation 同样获得相近的加速比。

Past two units, DGX Spark needs a 200G/400G switch. About $4,000 more. MaixStation needs a cable.超过两台,DGX Spark 需要 200G/400G 交换机,多花约 $4,000。MaixStation 只需要一根线。

Every tier is a cable away. **No expensive switch.**每一级只隔一根线。无需昂贵交换机。

Text‑to‑video and image‑to‑video run on your own box. No queue, no cloud bill, footage never leaves your desk.文生视频、图生视频跑在自己的机器上。不排队,没有云账单,素材不出你的桌面。

Point Claude Code, Codex or OpenClaw at the on‑device API. Flat cost.把 Claude Code、Codex、OpenClaw 指向设备上的 API,成本固定。

Real‑time policies at a fraction of Jetson‑class cost.策略实时执行,成本远低于 Jetson 级方案。

Full 27B on a power bank. Plug into any laptop.满血 27B,充电宝供电,插上笔记本就能用。

Agents for documents, mail and data. Add a second unit over MCIO when you want DeepSeek V4‑Flash 284B.处理文档、邮件和数据的 agent。想跑 DeepSeek V4‑Flash 284B 时,MCIO 加一台。

DeepSeek V4.1‑Flash 552B for the whole office. Stack boxes or cards as you grow.DeepSeek V4.1‑Flash 552B 服务全公司。一体机或算力卡按需级联。

── more in #ai-infrastructure 4 stories · sorted by recency
── more on @sipeed 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/sipeed-announces-mai…] indexed:0 read:2min 2026-10-15 · —