GitSeed 发布自托管 AI 编码智能体控制面
GitSeed 发布自托管、厂商中立的 AI 编码智能体控制面，把 Claude Code、Codex、Gemini、OpenCode 等运行时放进 prepare→agent→finalize 三阶段隔离容器中执行。
Why it matters: 原文完整交代了自托管控制面的三阶段容器隔离、签名回执与多模型并行评审，可据此判断智能体编码的治理路径。
Some English translations are not ready yet. The site is still in English. Untranslated news and event text is shown in its original language until background processing completes.
Technology areas
Deployment engineering Latest updates
Engineering practices for getting models running: inference optimization, VRAM and cost, serving architecture, and infrastructure selection.
74 条精选Past 30 days: 71 updatesTotal: 1,009 updates
Updated
Deployment engineering的精选
Hacker News AIFeaturedAI score7878
Hacker News AIFeaturedAI score7878 Anthropic upgrades Claude Code Projects to run parallel AI workers
Anthropic has updated Claude Code Projects so that users describe a goal and Claude scopes the request, delegates work to parallel threads, coordinates them, reviews outputs and assembles the result, with progress steerable from a phone. Each thread is a Claude Code cloud session on its own branch and repo copy, and threads now share a project memory and a library of added files and artifacts. The beta is available to select Claude Pro and Max subscribers using cloud sessions who have no existing web or desktop projects, with broader access over the coming week and Team and Enterprise plans after that.
Why it matters: It details a concrete shift in how Claude Code Projects delegate and coordinate multi-thread agentic work, plus the staged rollout limits.
Cloudflare BlogFeaturedAI score7171 Kitesurf is now open source
Cloudflare has open-sourced Kitesurf, its stateless browser for AI agents that runs entirely on Workers, under the Apache 2.0 license on GitHub. The repository includes the engine, PageScript, PageRenderer, and CLI, built on Rust-to-WebAssembly projects like Blitz, Stylo, Boa, and Linebender libraries, and supports the Chrome DevTools Protocol for clients such as Puppeteer. Developers can run it locally or deploy their own instance to Cloudflare Workers, which requires a Workers Paid plan because it uses Dynamic Workers.
Why it matters: Useful for developers who want to self-host a lightweight agent-oriented browser engine and inspect or contribute to its open-source components.
Claude BlogFeaturedAI score6262 Building always-on agents
This post explains that always-on agents are stateful and proactive, running on the same core loop as other agents but surrounded by architecture built from five primitives: an event inbox, timers, memory, permissions, and an outbox. It describes how each primitive works, including inbox best practices, follow-up versus scheduled timers, memory types such as user instructions and task progress, credential proxying and deferred tool calls, and outbox delivery pacing. It concludes with how the primitives form a repeating cycle in which each brief run is triggered by an event and can set timers that wake the agent again later.
Why it matters: Offers a concrete architectural blueprint for teams building proactive, stateful agents, with practical guidance on event handling, memory, permissions, and message pacing.
Hugging Face BlogFeaturedAI score6060 llama.cpp v1.0.0 : A stronger foundation for the local AI ecosystem
llama.cpp released libraries v1.0.0, marking a commitment to stable public APIs for downstream developers, with both llama.cpp and ggml now reflecting API and ABI incompatibilities through major-version bumps. The project states it supports more than 150 architecture variants and all mainstream speculative decoding methods, and distributes pre-release artifacts across Linux, macOS, Windows, Android and iOS plus channels including Docker, Homebrew, conda-forge, MacPorts, Winget and XCFramework. It also describes continued investment in CI infrastructure and self-hosted runners with hardware support from NVIDIA, AMD, Qualcomm, Arm, IBM and Intel.
Why it matters: Signals a versioning milestone for a widely depended-on local inference library, useful for developers tracking API stability and platform support.
vLLM BlogFeaturedAI score7272 DeepSeek-V4.1-Flash on Vera Rubin NVL72: 58x Throughput vs. GB300 NVL72 at 387 tokens/s
On a single NVIDIA Vera Rubin NVL72 rack, vLLM serves DeepSeek-V4.1-Flash at 1.7M tokens/s per GPU with 387 tokens/s per user, 58x GB300 NVL72, reaching 2.52M at peak throughput and 1,810 tokens/s for a single user. The gains come from kernels tuned for Rubin's split memory domains and FP4 tensor cores, specialized low-latency decode kernels, and a rack-wide KV cache and Engram tables shared over NVLink. The post details four optimization groups and reports a 96.5–97% prefix hit rate at high concurrency with workload-aware routing.
Why it matters: It offers a detailed, measurement-backed look at how Rubin-specific memory and MMA features translate into serving throughput and latency for a frontier open model.
vLLM BlogFeaturedAI score6060 Inkling 2 Small on vLLM: Day-0 Support
vLLM announces day-0 support for Thinking Machines' Inkling 2 Small, a multimodal text-and-vision MoE model with 594B total and 17B active parameters, up to 1M tokens of context, and a native DSpark speculator. The post details the architecture (48 layers, 384 experts with top-8 routing, block attention residuals, gated attention, a 23-layer ViT image encoder, and mixed NVFP4/MXFP8 expert precision) and vLLM's optimizations, including kernel fusion with PDL, fused MoE activation quantization, sequence parallelism for prefill, and a Gluon relative attention kernel, reaching up to 630 TPS/user at TP4 on GB200. It also covers LoRA and RL (via SkyRL) support, a serve command, agentic-workload benchmarks, and accuracy checks on NIAH up to 1M tokens, MMMU-Pro and HLE.
Why it matters: Useful for engineers evaluating or deploying a large multimodal MoE model, as it pairs concrete architecture and quantization details with vLLM-specific kernels, benchmark numbers and a ready serve command.
AWS Machine Learning BlogFeaturedAI score7171 Run agents securely and efficiently with the next generation AgentCore runtime
AWS announced updates to the Amazon Bedrock AgentCore runtime that keep each session in its own hardware-isolated Firecracker MicroVM while starting sessions up to 15x faster, scaling to 4x more compute (up to 8 vCPUs and 32 GB, with container images up to 10 GB), and running up to 68% more efficiently. The runtime now reclaims memory within minutes of release and scales idle agents to zero, and agents can dynamically mount and save work to a filesystem such as an Amazon S3 bucket with FUSE support so data stays inside the trust boundary. The next generation runtime is available today in AWS Regions across North America, South America, Europe, and Asia Pacific.
Why it matters: Announces concrete performance, scaling, billing, and data-isolation changes to a managed agent runtime, with customer metrics and availability details useful to teams planning production agent deployments.
Hacker News AIFeaturedAI score7878 Vercel Data Shows AI Models Have Six-Week Half-Life in Production
Vercel's November 2026 AI Gateway Production Index reports that production AI models have an effective half-life of about six weeks, with more than half of October's language-model requests going to models under 60 days old and most teams abandoning a model within four weeks of its peak. Decision models reached 26% of gateway requests in October at just 0.53% of spend, voice usage rose from 4% of paid teams in July to 11% in October, and Anthropic and OpenAI were nearly level at 54 and 55 of every 100 paid teams. Image and video spending hit records, with OpenAI's GPT-Image-2.5 Sunburst leading image spend and ByteDance's Seedance 2.5 leading video spend for a third month.
Why it matters: The index offers a rare production-usage view of model churn, spend shifts and emerging decision and voice workloads, useful for teams planning model-swappable pipelines.
vLLM BlogFeaturedAI score6363 H3-Image Day-0 Support: 4K in 0.87s with Vera Rubin
vLLM-Omni brought MiniMax H3-Image, an open-weight text-to-image and editing model, to NVIDIA Vera Rubin, reaching 0.87 seconds median for native 4096×4096 generation on four GPUs with SP4 and MXFP8 in a short-prompt serving benchmark. A separate SP1×DP4 throughput configuration delivers 6,115 images/hour, which the authors project to 146,755 images/day and, at an assumed USD 0.06879 per image, USD 105.16/GPU/hour gross revenue at full billable utilization before operating costs. The post also documents VAE decoding gains from CUDA Graph replay plus multiple streams (2.635 s to 0.213 s on four GB300 GPUs) and community validations on AMD, Intel, Huawei Ascend and Moore Threads hardware.
Why it matters: Useful for readers weighing open-weight 4K image serving, since it separates latency and throughput operating points and labels its revenue projections as extrapolations before operating costs.
Google Blog AIFeaturedAI score8080 Gemini Decisions API: Fast, calibrated multimodal decisions
Google introduced the Gemini Decisions API, powered by the new Gemini Decision 1 model (gemini-decision-1) and available through the Gemini API in Google AI Studio, for closed-set decisions such as tool selection, policy checks, model routing, and context grading. It accepts one multimodal input (text, images, audio, video, PDFs up to 256K tokens) with up to 1,000 yes/no, multiple-choice, and rubric-scoring questions, scoring options in a single forward pass with confidence scores at 150-250 ms end-to-end latency and $0.10 per 1M input tokens with zero output tokens billed. Google reports that Gemini Decision 1 leads across agent routing, reasoning, ranking, and multimodal benchmarks with the lowest calibration error, and is nearly 5x cheaper than Gemini 3.5 Flash-Lite with the Interactions API across the shared benchmark suite.
Why it matters: It documents a new Google model and API with pricing, latency, and benchmark claims, useful for teams weighing low-latency classification and routing options.
ElevenLabs BlogFeaturedAI score6060 ElevenLabs and NVIDIA expand collaboration to bring voice agents to enterprises and governments globally
ElevenLabs and NVIDIA are expanding their partnership across models, infrastructure and sovereign deployments, with NVIDIA Nemotron 3.5 Super VL now available in ElevenAgents and ElevenLabs adding 15MW of NVIDIA Blackwell Ultra and Vera Rubin compute. ElevenLabs models and NVIDIA infrastructure power a sovereign voice agent for Poland's national e-Health appointment system, and the companies are exploring on-device voice AI for robotics and physical AI.
Why it matters: It shows how voice-agent stacks are being bundled with model and compute partnerships, and includes a concrete sovereign healthcare deployment with stated operational details.
AWS Machine Learning BlogFeaturedAI score6060 UAEF: A framework-agnostic system for closed-loop evaluation of AI agents
AWS introduces the Universal Agent Evaluation Framework (UAEF), an open-source, framework-agnostic system that uses adapters to normalize agent traces from LangGraph, LangChain, Strands Agents, Amazon Bedrock AgentCore, and Langfuse into a common AgentTrace format, then scores them with 36 built-in metrics across 7 quality dimensions. Results are stored per experiment in Amazon DynamoDB and Amazon S3, enabling comparison of agent versions, regression detection, and report and insight generation, with both a Python library and a low-code web UI sharing the same engine. The post also provides a playbook for choosing metrics and preparing datasets, plus runnable notebooks for each supported framework.
Why it matters: It offers a detailed, practical reference for teams evaluating agents across frameworks, including metric selection guidance and a closed-loop workflow for catching regressions before release.
Databricks BlogFeaturedAI score7171 Genie One: The AI cowork home for everyone
Databricks announced that Genie One is now the enterprise home for AI cowork, combining Genie Search (hybrid AI search indexing tools like Google Drive, Jira, Confluence, Workday and SharePoint, plus over 100 MCP connectors), organizational directories, in-conversation artifacts such as slide decks and dashboards, MCP-based actions across tools, shareable skills, and native apps for browser, desktop, mobile, Slack and Microsoft Teams. Genie can now take more autonomous action on users' behalf, with Unity Gateway providing traces, MCP policies, permissions and cost management, and Genie One is free for users through January 31, 2027.
Why it matters: It details a major enterprise AI assistant release, useful for readers tracking how agentic workplace tools are being packaged with governance and cross-tool search.
Databricks BlogFeaturedAI score6969 Announcing GA of Genie Ontology
Databricks announced the General Availability of Genie Ontology, an agentic ontology layer that agents build by learning from dashboards, SQL queries, notebooks and BI semantic models, while people govern definitions through Unity Catalog semantics. The release adds Genie Search (GA), Ontology Snippets (GA), external knowledge sources (Beta, starting with Power BI), organizational directories (GA, starting with Workday), Domains and Pages (GA), and Ontology Manager (coming soon in Beta). In a benchmark built with Turing simulating a Databricks workspace for a fictional UK energy supplier, Genie answered correctly in 91% of runs versus 45% and 58% for two frontier agents, using less than half the tokens and finishing three to four times faster.
Why it matters: It details a major enterprise context-layer release and includes benchmark numbers comparing agents with and without the ontology.
Databricks BlogFeaturedAI score7474 The new Genie: autonomous AI for complex enterprise work
Databricks introduces the new Genie, an autonomous AI coworker that continuously handles long-running enterprise work, tracking progress, following up on missing inputs, and taking action as new information arrives, with users setting guardrails and approving where needed. It is supported by Genie Ontology for business context, Genie Search for hybrid enterprise search, Unity Gateway and Unity Catalog for governance, and new capabilities including artifacts, MCP connectors, Genie Agents, and plugins for tools such as ChatGPT and Claude. Databricks reports that in a new benchmark built with Turing, Genie answered correctly in 91% of task runs, 1.6 times the rate of the stronger frontier agent, while using less than half as many tokens.
Why it matters: The announcement details a shift from task-level AI assistance to supervised autonomous agents, with concrete governance, context, and benchmark claims worth evaluating.
The DecoderFeaturedAI score8282 Anthropic says AI is eroding Nvidia's CUDA lead, cutting an AMD chip switch from months to three days
Anthropic claims it ported its Claude model to AMD GPUs in about three days with almost no human intervention, using its own frontier model, Mythos, to write or port nearly 100 percent of the kernels, versus the six to twelve months traditional engineering approaches require. The company says it now trains and runs Claude across four chip families—AWS Trainium, Google TPUs, Nvidia GPUs and AMD GPUs—treated as one interchangeable compute pool, and it cites figures from an investor presentation that have not been independently verified. Anthropic also points to "compute multipliers" and its "Anthropic Direct" data center program, while Nvidia and AMD both hold financial ties to the company.
Why it matters: It offers a concrete, if unverified, data point on whether AI-assisted porting can loosen CUDA's lock-in and shift how AI labs choose among chip vendors.
OpenRouter AnnouncementsFeaturedAI score6767 MCP vs CLI vs Skills: What Each Costs in Context
The article measures input-token costs for giving an agent GitHub access via the GitHub MCP server, the gh CLI, or a gh skill across Claude Haiku 5.5, Claude Sonnet 5.5, GPT-6 Luna, and Gemini 3.8 Flash; on Sonnet 5.5 the CLI used 1,421 tokens, the skill 3,315, and the MCP server 59,417 with its 46 default tools. It reports that the GitHub MCP server's 46 default tools add 19,436 tokens to every request on Sonnet 5.5, and that a single tool result (GitHub's release object) cost about 20,400 tokens. It also describes the beta Tool Search server tool, which cut the up-front cost of 46 tools to under 600 tokens and halved whole-task cost on Sonnet 5.5 while raising it on Haiku 5.5.
Why it matters: It offers concrete, model-specific token and cost measurements for choosing between MCP, CLI, and skills, plus practical mitigations like scoping tools and trimming results.
PyTorch BlogFeaturedAI score7171 ExecuTorch WebGPU: PyTorch Models on the GPU, in the Browser
ExecuTorch's new WebGPU backend runs PyTorch models exported with torch.export to a .pte file directly on the GPU in WebGPU-enabled browsers, without converting to another framework's format. The same artifact runs across Chrome and Safari on Metal, Vulkan and D3D12, and on an Apple M1 Pro in Chrome it was 4–7× faster than ExecuTorch's Wasm (XNNPACK) CPU path on prefill and 2–10× faster on decode. Validated models include Llama 3.2, Qwen3, Gemma 3, SmolVLM, Depth Anything V2, SAM 2, Whisper and on-device LoRA fine-tuning, all running locally in the tab.
Why it matters: It documents a concrete path for running PyTorch models on browser GPUs with benchmark numbers and a list of validated models, useful for developers weighing in-browser inference.
Cursor BlogFeaturedAI score7171 Cursor Projects is now multiplayer
Cursor announced that Projects, its multi-agent workspace introduced in September, now supports multiplayer so teammates can plan, build and ship together alongside their agents. Projects adds a shared context system holding requirements docs, diagrams, tech specs and mockups, plus a shared task board showing who and which agents are working on what and the PR behind each task. The feature also enables agent-to-agent communication across teammates, and is rolling out in beta to all Cursor users starting today.
Why it matters: Useful for teams evaluating collaborative AI coding tools, as it describes a shared-context multi-agent workspace and its beta availability.