クラウド

Kern 是一个 1.5MB 单二进制、无守护进程的容器与资源运行时，基于 OCI 镜像、cgroup v2 和 namespace，启动约 3.5ms。

I just finished building my (relatively) low rent local inference machine: * Epyc 7663 * 256GB ECC DDR4-3200 * 1x RTX 5090 32GB Yeah I realize it's weird to throw a 5090 and 256GB of anything together and call it low end, but relative to ~151GB of weights it is. I'm running UD-Q8_K_XL and getting 23.8-24.6 tokens/sec, with pp ranging from 60 on the first prompt to 385 near the last (no doubt lots of caching) on tasks using 100-128k total context. It was slower with DFlash so I took that out. It

PicoMQ 是基于对象存储的持久流服务器，通过 HTTP 提供 URL 可寻址的流，支持 long-poll 与 SSE。

arXiv:2608.20397v1 Announce Type: new Abstract: Agentic large language models (LLMs) on the Model Context Protocol (MCP) re-encode verbose tool schemas every turn, so prefill - quadratic in sequence length - dominates time-to-first-token (TTFT) as the tool registry grows. Nexus's primary lever is to decouple routing from the schema-prefill cost: an INT8 semantic lookaside buffer (SLB) with a calibrated cross-encoder margin gate selects tools by retrieval, and arguments are generated over a compr

The five largest GPU neoclouds now run on very different models. CoreWeave and Nebius report to the SEC; Lambda and Crusoe are private and heading toward IPOs; Groq rebuilt itself as an inference cloud after licensing its LPU technology to NVIDIA. This comparison checks each provider's live rate card, Q2 2026 financials, active and contracted gigawatts, anchor contracts, and SemiAnalysis ClusterMAX tier. Nebius posts the lowest H100 rate and the only published B300 price, Lambda has the cheapest

Credit to Twitter Post

As the memory shortage continues to cause trouble for hardware makers, Amazon says it is now being forced to pass on the costs to its consumers.

FreeToken splits MoE cache misses between PCIe fills and CPU execution using measured bandwidths, unlocking frontier models locally The post Meet FreeToken: An Edge-Native MoE Serving Engine that Runs 753B GLM-5.2 on a Single Workstation GPU appeared first on MarkTechPost .

开发者用 8 张 B300 托管 Kimi K3（2.8T 参数），MXFP4 下解码 92 tok/s，成本约 190 美元/百万输出 token。

用户将自建 DGX Spark 集群从 16 节点扩至 36 节点，总统一内存达 4.6TB，用于本地运行各类大模型。

it’s powerful and quiet. I run all of my services on it. Specs: AMD Ryzen 7 7840, 32 GB of RAM, 2 TB SSD. idle power consumption is about 5 watts. Additionally, it’s very tiny. I’ve taken it with me to 4 countries in recent years.

自托管社区讨论是否仍有人坚持纯 Linux + Docker Compose 方案，多数人认可其简单可控，但 UI 工具仍有存在价值。

The Dutch Data Protection Authority is fining Uber €825 million in the second largest penalty issued under Europe’s GDPR.

I see so many people in here using Proxmox. I run Ubuntu + Docker, and my whole setup is just a list of folders with compose.yamls and supporting files (configs etc). It feels very manageable this way, fully declarative, easy to version control. Am I missing something, that makes people recommend and use Proxmox instead?

Dan Luu 撰文称软件没有理由再慢下去，LLM 已将性能优化的成本降低数个数量级。

评论普遍认为软件变慢源于优先级、激励和工程权衡，而非技术能力；但也有人认为AI优化有限，慢软件仍将存在。

It was only a matter of time...

A Dutch data regulatory authority said that Uber has to pay 824.9 million euros for violating the GDPR.

Cheap energy, abundant land, and proximity to Beijing have turned a city in Inner Mongolia into a crucial hub for data centers.

Has anyone else noticed an increase in scanners/bots in the past ~month? For the past couple years I've had 2-3k hits a day from bots but lately there has been a steady increase in traffic looking mostly for php files. What I find strange is how much of this traffic is coming from MS and Google IPs. Do they not have any kind of monitoring on their cloud services? Having thousands of requests spamming every IP that responds should raise some flags. 20.24.67.246 Hong Kong Hong Kong Microsoft Corpo

Encore 在 Apple Silicon 上重建 Linux MicroVM 栈，让 Mac 开发者能在本地运行 Firecracker 微虚拟机。

Run Claude Code, Codex or Opencode in cloud with your repo Discussion | Link

Nvidia continues to pour money into data center development — just as AI data centers bring lots of money into Nvidia.

Yes, we’re confused too.

Component Validated configuration Motherboard ASRock Rack SPC621D8U-2T/OVH CPU Xeon Gold 6330 (Get gold/platinum if interested in Optane Pmem gimmicks) GPU fabric Two Broadcom/PLX PEX88096 islands, eight GPUs per island GPUs 16 x RTX 5060 Ti 16 GB OS Ubuntu 22.04.5 LTS Kernel 6.8.0-106-generic NVIDIA driver Aikitoria patched open driver 610.43.02-p2p Required BAR1 16,384 MiB on every GPU UEFI boot enabled; CSM disabled. Secure Boot disabled. The locally built EFI application and patched NVIDIA m

I just started my self-hosting journey in April after I got tired of my Xfinity modem. I ran what feels like miles of Cat 6, built a rack, bought the PCs and domain, etc etc and now I feel super accomplished. I knew practically nothing about all this back in March. Possibly the most fun I’ve had is building this simple little homepage dashboard to give everything a face. I stole lots of inspiration from everyone here. It now serves as a true homepage with all the frequent sites and services I ev

The new traffic peak caused "capacity failures" that led to the developer platform's recent outage.

FreeToken 是边缘原生 MoE 推理系统，将个人机器视为弹性推理平台，动态调度 CPU-GPU 与专家驻留。

🤖FreeToken is an edge-native Mixture-of-Experts serving system that dynamically maps computation and model state onto heterogeneous local hardware to run large open-weight models on personal machines.

The AI buildout shows no signs of slowing. And with hundreds of billions of dollars a year going into data centers and GPUs, compute has become the single biggest cost for anyone building AI products. But for all that spending, there still isn’t a straightforward way to put a price on compute — or for firms to hedge their exposure when the price changes. Silicon Data […]

Previously: Stripe will reportedly acquire OpenRouter for $7B+

多数评论祝贺OpenRouter被收购，但担忧企业整合损害用户体验和数据安全；也有人认为Stripe可借此构建AI计费基础设施。

the Memory crunch continues - Moore’s Law reversed to 2007 levels

arXiv:2608.16956v1 Announce Type: new Abstract: API buyers purchase a dated contract, not a model name alone: the contract includes the requested and served model, reasoning-effort term or its omission, output rail, service product, prompt, and price schedule. We study the reasoning-effort term through a registered paired contrast of Sonnet 5 with explicit high effort against the same model with effort omitted, using 30 AIME 2026 items and five calls per item. Every paid attempt was assigned one

Cerebras CS-4 宣称推理速度比 GPU 快 30 倍，评论区认可性能但质疑功耗、价格与可及性。

评论区普遍认可Cerebras CS-4推理性能强劲，但质疑其功耗、价格、模型更新及可及性；也有人认为其将挑战英伟达。

《PostgreSQL for Everything》主张 Postgres 作为通用默认数据库，评论区认可其地位但也指出队列、分析等场景的局限。

多数人认可PostgreSQL作为通用默认选择，但也有人认为SQLite或MySQL更简单，且PostgreSQL在队列、分析等场景有局限。

Bankrupt Spirit accused of selling out workers in massive data sale to Google.

TerraPower's nuclear power plant possesses a strategic advantage over competitors, especially when chasing after data center deals.

On Tuesday, Warp introduced Warp Factories, a new infrastructure system designed to make building AI software factories as easy as possible.

Large language model (LLM) agents are evolving from conversational assistants into autonomous systems that execute long-horizon tasks through reasoning, tool use, code generation, and workspace manipulation. As agents increasingly operate over persistent environments and multi-step workflows, they face challenges analogous to those addressed by transactional database systems: reliable execution, consistent outcomes, safe concurrency, and durable state management. We introduce the concept of an a

🤖An ACID-compliant framework for agentic transactions introduces semantic guarantees to ensure reliable, isolated, and durable execution of long-horizon LLM agent workflows.

Glean CEO Arvind Jain explains why model routing helps control AI costs for organizations, and how human feedback loops at scale improve its routing systems.

changed a db password in my .env, ran docker restart on the stack, then spent an hour convinced the db was corrupted because auth kept failing. turns out restart just brings the container back with the exact config it was created with. env is only read at creation. docker compose up -d --force-recreate fixed it in ten seconds. two years running this stack and never got bitten by it until now. what's the dumbest thing that ate an evening for you?

No GPUs, no Agents, just really, really, really good infra and distribution.

Linux 7.3 合并 VRAM 超限性能优化补丁，评论区赞赏内核改进，但也有人关注 OOM 崩溃与 NVIDIA 支持问题。

评论区普遍赞赏Linux内核优化VRAM性能，期待改进落地；但也有人认为需关注OOM崩溃和NVIDIA支持问题。

This is the latest large-scale DDoS attack to hit the social networking site this year.

A few days ago I had the slightly ridiculous idea of building a mobile homelab inside my everyday backpack. Today I started looking through some old hardware I had lying around and realized: I can actually build this. And I barely need to buy anything. 😂 The main machine is an old ASUS TUF Gaming F15 that I bought from a friend for €300 and barely used: Intel Core i5-10300H 64 GB RAM GTX 1650 2.5 TB SSD storage across three physical drives Built-in battery = basically a tiny UPS Pop!_OS + Window