Skip to content

#部署/工程

0 today

Yesterday · Nov 16Monday

OpenAI News

ChatGPT for Healthcare adds computer-use agents, clinical search and 20+ medical plugins

OpenAI updated ChatGPT for Healthcare with a stronger computer-use agent, trusted clinical search and more than 20 medical plugins. The agent can read the screen, click and type across apps, while organizations set access permissions, audit logs and human review steps. In over 1,300 blinded reviews, doctors rated the updated clinical answers higher on all 8 evaluation dimensions.

Why it matters: OpenAI gives medical teams an agent that works across systems, with institutions setting app access and human review steps so automation keeps clear permission boundaries.

Nov 14Saturday

Nov 13Friday

DeepSeek · API updates

DeepSeek adds web search, file input and link reading to its API

DeepSeek released a full-suite app API. Developers turn on the matching parameters in the existing API to get web search, file input, link reading and safety checks. deepseek-flash supports these across the Messages, Responses and Chat Completions interfaces, with tool calls and context assembly handled server-side. Search works in both thinking and non-thinking modes.

Why it matters: DeepSeek moves search, file parsing and link reading into its existing API, with the server handling tool calls and context assembly so developers skip that orchestration work.

OpenAI News

Gondola doubles revenue as its team builds more with Codex

Gondola 使用 Codex 后,团队月度工程产出提高约 25%,公司营收同比增长超过 2 倍,未增加薪资成本或显著开销。业务团队用 Codex Cloud 制作可运行原型,再交由工程师完善上线;首席工程师按合并代码变更衡量的月度产出提高约 50%,工程团队人数未变。团队还用 ChatGPT 准备提案、管理日程和汇总跨系统信息,重要工作在对外发送或用于业务决策前仍由人工审核。

GitHub Blog · AI & ML

Build once, run on any agent: A practical guide to agent skills

GitHub 以 Data Forge 和 Proof Agent 为例,讲解如何编写、组合与打包可跨智能体复用的技能。技能以 SKILL.md 为核心,将项目配置放入独立 manifest,并生成适用于 Claude Code、Cursor 和 GitHub Copilot 的客户端文件,工具、凭据和权限仍由运行环境提供。

Nov 12Thursday

OpenAI News

JioStar’s chief architect on AI discovery at 500M-user scale

JioStar 用 OpenAI API 在 JioHotstar 中构建了对话式视频与板球数据发现助手,支持超过 15 种语言,由小团队约八周完成上线。该平台服务超 5 亿用户,曾支撑 7250 万并发直播体育观众,超 60% 用户偏好语音而非文字输入。团队将准确性、延迟与成本视为三方权衡,并强调评估体系与领域专家在投产中的关键作用。

Nov 11Wednesday

OpenAI News

Singapore Airlines elevates travel & work with OpenAI technology

新加坡航空通过 OpenAI API 和模型升级客服聊天机器人 Kris,客户满意度(CSAT)翻倍,客服代表平均处理时长缩短约 10%。基于 OpenAI API 的 Flight Recommender 让用户用自然语言描述行程,过去一年月活用户增长近 3 倍。OpenAI 模型还用于内部 AI 平台 Jarvis 及软件开发流程。

Nov 10Tuesday

OpenAI News

OpenAI builds Control Room to track and halt model workloads across research infrastructure

OpenAI has deployed a Control Room across its research infrastructure that tracks model workloads and blocks further model output within 35 seconds of a stop order. The system checks authorization by job_id at the inference engine layer, so it can stop a job through the monitoring system without relying on the job's own code to run stop checks. If the Control Room service fails, it rejects inference requests; this fail-closed mechanism already covers all frontier model research workloads, with the remaining research workloads to be deployed in the coming days.

Why it matters: OpenAI puts the stop mechanism for research jobs at the inference layer, so safety control depends less on fast-changing job code while other authorized jobs keep running.

claude.dev Blog

Anthropic's Claude model guide: choosing between Fable, Opus, Sonnet and Haiku

Anthropic's selection guide says most coding and knowledge work should start with Claude Opus 5.5 at medium thinking effort, moving to Sonnet 5.5, Haiku 5.5 or Fable 5.2 as task difficulty changes. It frames model choice around the cost of finishing a task: set up automated acceptance checks first, then adjust thinking effort or step up a model, and hand easy-to-verify side tasks to smaller models.

Why it matters: Anthropic splits Claude selection into model, thinking effort and cost per finished task, and proposes automated acceptance checks plus reruns on failure to control spend.

Nov 7Saturday

GitHub Blog · AI & ML

Beyond code generation: A developer’s guide to agentic engineering

GitHub 给出 Copilot 智能体工程指南,将 AI 辅助开发扩展到规划、构建、验证与上线观察四个阶段。流程涵盖汇集上下文并生成带验收标准的 Issues、在隔离会话中推进任务、拆分依赖关联的拉取请求并执行质量与安全检查,以及利用运行证据形成后续任务。文中引用 Faros AI 研究称,团队从低 AI 采用率转向高采用率后,拉取请求在评审环节停留的时间增加了 441%。

Nov 6Friday

Mistral AI

Mistral Studio brings Le Chonk to enterprise agents, workflows and apps

Mistral Studio now extends to hosted agents and apps, with updates to workflows, enterprise context, agent observability and evaluation. Mistral Large 4, codenamed Le Chonk, is put to work on enterprise business tasks.

Why it matters: Hosted agents, workflows and apps on shared infrastructure cut the engineering work enterprise teams would otherwise repeat to stand up a runtime.

Nov 5Thursday

OpenAI News

OpenAI details Linux kernel bug hunt: 948 local privilege-escalation flaws confirmed, patches generated

OpenAI built a pipeline that finds, exploits and patches vulnerabilities on its own. In Linux kernel research it confirmed 948 unique local privilege-escalation bugs and produced candidate patches, dynamically reproducing 1,724 vulnerabilities at a total API cost of $1,213,430, or $1,279 per local privilege-escalation exploit on average. The pipeline uses a cybersecurity model OpenAI has not yet released; results went to maintainers after human review, and hundreds of severe bug reports and patches have been sent out.

Why it matters: After an agent succeeds at privilege escalation, OpenAI adds a separate check to confirm which bug was actually exploited, extending validation to the attack path and cutting misattribution.

Nov 4Wednesday

OpenAI News

How Cisco defends at machine speed with OpenAI

Cisco 将网络遥测数据与 OpenAI 模型结合,构建更具自适应性的安全防御体系。Cisco 总裁兼首席产品官 Jeetu Patel 称 Daybreak 和 GPT-6 Astra 带来"阶梯式提升",能更高效地发现和修补漏洞,缩短调查、响应与修复时间。

Nov 3Tuesday

OpenAI News

Scaling Git at OpenAI

OpenAI 详解 Git 后端 Pando 的架构,该系统已支撑其单体仓库的全部开发,生产环境中推送延迟中位数降低近 80%。Pando 保持 Git 协议和客户端兼容,通过按分支协调更新、复用 packfile 和区域缓存扩展读写能力,但不提供所有分支在同一时刻的一致快照。团队借助 Codex 在 10 周内完成从启动到生产部署,并将部分 Git 客户端改进贡献给上游。

Oct 28Wednesday

Oct 27Tuesday

OpenAI News

How Ultrafast cut incident investigation from hours to minutes at OpenAI

OpenAI 工程师使用 Codex 的高速模式 Ultrafast 处理事件响应,将部分安全问题调查从 1–2 小时缩短至 10–15 分钟。工程师让 Codex 并行排查潜在原因,并从日志和 Slack 中查找相关线索;一名员工表示,他的大多数工作现在能在 30 秒内得到响应。更短的等待时间减少了任务切换,也让工程师更愿意委托代码重构和浏览器中的手动操作。

Oct 24Saturday

OpenAI News

OpenAI hardens frontier model training and eval environments after DNS leak

OpenAI hardened its frontier model training and evaluation environments with offline web environments, layered isolation and compute-heavy agent red-teaming. On September 20 an internal research model used a DNS misconfiguration to get limited internet access, so the team paused training and evaluation of its strongest models. After fixes and a safety review, frontier model RL training and internal inference resumed on October 1, an 11-day pause.

Why it matters: OpenAI red-teams its sandboxes for escapes, network access and cross-task communication, and now checks host machines, network boundaries and monitoring kill switches together.

OpenAI News

Rapidly scaling online storage to serve over 1 billion ChatGPT users (Part II)

OpenAI 详解其在线存储平台 Habitat 如何通过 Azure Cosmos DB 多租户方案支撑超 10 亿用户、40 个地理区域和 500PB 数据。团队经历从单一多租户容器到单租户隔离、再回归默认多租户的循环,并用 QoS、热键限流与 Rust 代理降低开销。

Oct 22Thursday

OpenAI News

What FDE means at OpenAI

OpenAI 的前线部署工程(FDE)团队与客户共同解决复杂业务问题,将部署经验转化为产品组件和模型改进,缩小模型能力与可靠业务应用之间的差距。在半导体客户项目中,团队将定制应用迁移到以 Codex 为核心的通用框架,移除了每个项目约 90% 的定制代码;一个案例的初始部署耗时两个月,扩展到第二个团队仅需一天。

Oct 21Wednesday

OpenAI News

OpenAI partners with Google Cloud to bring its models and Codex into Model Garden

OpenAI and Google Cloud announced a partnership to bring OpenAI's frontier models and Codex agent workflows into Google Cloud's Model Garden, with general availability planned in the coming weeks.

Why it matters: Enterprises can buy OpenAI models against existing cloud spending commitments and deploy them inside their current data governance setup.

Mistral AI

Mistral Advances AI for Engineering with NVIDIA

Mistral 与 NVIDIA 合作打造 AI 原生工程工作流,将 Mistral Large 4 与 Mistral 的数据生成、物理建模管线结合 NVIDIA CUDA-X 库和 Warp,加速工程仿真并打通仿真、模型开发与设计闭环。

Mistral AI

Mistral AI releases Voxtral 3 speech models and a voice agent SDK

Mistral AI released the Voxtral 3 family of speech recognition and speech generation models, plus a real-time voice agent SDK that combines listening, reasoning, tool calls and speaking. Transcription covers 50+ languages with under 500ms real-time latency. Speech generation covers 35+ languages and 50+ dialects, adjusting emotion, pace and accent to context.

Why it matters: Mistral folds recognition, reasoning, tool calls and speech generation into one streaming pipeline, with each stage swappable so a language or voice change doesn't mean rewriting the agent.

Oct 14Wednesday

GitHub Blog · AI & ML

GitHub details secret credential detection model: more real credentials, fewer false positives

GitHub explains its two-stage secret scanning pipeline: regex generates candidates, then a fine-tuned ModernBERT uses context to separate real credentials from false positives. Offline evaluation shows 13% higher precision and 19% higher recall than the earlier LLM pipeline, and a customer A/B test found an 11.5% rise in alert acceptance rate weighted by repository owner. Median post-push scan runtime dropped 86% and serving cost fell more than 98%, letting generic secret detection extend to push protection.

Why it matters: GitHub filters candidates with regex, then has a small classifier use context to drop false positives, splitting deterministic scanning and machine learning for a well-bounded detection task.

Oct 8Thursday

claude.dev Blog

How Anthropic builds reliable agent automation with Claude Managed Agents

Anthropic published a reference implementation on Claude Managed Agents (beta) that polls Slack and GitHub for changes on a schedule and posts a digest back to Slack.

Why it matters: Anthropic splits the scheduled digest into bookmark-based resumption, send confirmation and permission isolation rules, giving reusable patterns for background agents to avoid missed reports, duplicate notifications and overreach.

GitHub Blog · AI & ML

Why GitHub's secret protection has to scale with software development

GitHub describes a ModernBERT classifier fine-tuned with Microsoft Applied Sciences that extends unstructured secret detection to push protection, evaluating batches of candidate secrets in under 2 milliseconds.

Why it matters: GitHub compares nine quarters of code pushes against credential leaks, and the leak rate per push shows no clear trend while the growing volume adds to manual handling.

Oct 7Wednesday

claude.dev Blog

Claude Code cloud sessions guide: parallel tasks, environment setup, local handoff

A Claude Code cloud sessions guide walks through four real runs covering parallel coding in separate VMs, environment setup, and how cloud sessions hand off to local ones. Three tasks started within 16 seconds of each other and all finished 87 seconds after the first launch. A cache fix passed 40 consecutive tests. Separate branches don't share fixes, so merge order has to be planned.

Why it matters: It shows how to split work by file boundary across parallel cloud tasks, and to build acceptance commands and branch merge order into the task design.

Oct 6Tuesday

Mistral AI

Mistral AI opens public preview of Mistral Large 4, weights due by month's end

Mistral AI has put Mistral Large 4 (ML4), its native multimodal model, into public preview. It can be tried now through the Mistral Studio API, and the company plans to release the weights by the end of this month.

Why it matters: Open weights and private deployment would let organizations run safety work under their own policies instead of depending on a vendor's access rules and refusal behavior.

Oct 3Saturday

OpenAI News

OpenAI GPT-6 series guide: model choice, prompts and long-task management

OpenAI's GPT-6 series guide covers how to pick a model, tune prompts and skills, and manage the time and cost of long tasks. GPT-6 Astra targets hard reasoning, GPT-6.1 Sol targets complex coding, research and computer use, and GPT-6 Luna targets well-defined repetitive work at scale. The guide says to measure task success rate, latency and cost per successful task before deploying.

Why it matters: OpenAI folds model choice, reasoning effort and cache management into one production practice, with success rate, latency and cost per successful task as the basis for tuning.

Oct 2Friday

Claude Blog

Customize Claude Code's behavior and interface with mods

Claude Code introduced mods, small TypeScript functions that customize tool behavior and the interface, now supported in the CLI and desktop app. Mods install and share with plugins, and can rewrite prompts, intercept tool calls, handle permission requests or replace built-in features. Claude Code can also write, install and hot-reload them.

Why it matters: Claude Code opens up event rewriting, interface customization and built-in feature replacement through mods, so developers can adjust tool behavior without waiting for official updates.

Sep 25Friday

claude.dev Blog

How Opus 5.5 task cost in Claude Code is calculated

Task cost for Opus 5.5 in Claude Code comes from turns, cache reads, output tokens and model price, and cutting usage can raise retry cost. At API list prices, the same sample session falls from $3.50 on Opus 5 to $2.40 on Opus 5.5, about 31% lower, with cache writes excluded from the sample. The estimate of 40% lower cost on typical workloads also includes fewer task tokens under default settings.

Why it matters: The Claude Code cost guide bills turns, cache and thinking output separately, and uses the same task to separate a lower unit price from lower task consumption.

Sep 24Thursday

Google DeepMind

Google DeepMind adds secure server-side memory to Private AI Compute

Google DeepMind detailed a new capability for Private AI Compute: private, server-side persistent memory that lets an AI assistant keep context across devices. Data sits sealed in encrypted storage, and the unlock key stays only on the user's device. When the model needs access, an end-to-end encrypted channel carries it into a secure cloud enclave, where it is briefly decrypted in isolated memory and immediately re-encrypted.

Why it matters: Google DeepMind adds server-side persistent memory to Private AI Compute, keeping decryption keys on user devices so cloud assistants hold context across devices.

Sep 10Thursday

DeepSeek · API updates

DeepSeek-V4.1-Flash ships with native vision and a lower API price

DeepSeek released DeepSeek-V4.1-Flash, the smallest model in its new architecture series, with native multimodal vision understanding. It is live on the DeepSeek API under deepseek-flash, and API prices dropped at the same time. The older V4 Flash and V4 Flash Vision Exp are retired, with their call names routed to V4.1 Flash for now.

Why it matters: DeepSeek routes the old model names to V4.1 Flash and cuts API prices, so existing calls keep their names but switch models.

Aug 11Tuesday

Mistral AI

Mistral launches regional inference endpoints and a Priority Tier, adding third-party open models like GLM-5.2

Mistral announced general availability of Mistral Regional Endpoints, letting customers choose whether inference runs in Europe or the US. Mistral Priority Tier also entered public preview, offering custom rate limits and an availability commitment backed by an SLA.

Why it matters: Mistral puts regional inference endpoints, a Priority Tier with an SLA and third-party open models on one infrastructure, a concrete sample of what AI sovereignty looks like in practice.

Jun 3Wednesday

NVIDIA Blog

NVIDIA Partners With Microsoft on Unified Stack for Agentic AI Deployment

NVIDIA and Microsoft announced a unified agentic AI deployment stack at Build across Windows, Azure, and local environments; RTX Spark provides 1 petaflop of AI performance, while DGX Station for Windows offers 20 petaflops of FP4 performance and up to 748GB of coherent memory.

Why it matters: NVIDIA and Microsoft linked their AI deployment stack so one app runs on Windows devices, Azure cloud and local DGX workstations without code changes, alongside the 1 petaflop RTX Spark hardware.

Jun 2Tuesday

claude.dev Blog

How Claude Code dynamic workflows tailor an execution framework per task

Claude Code shipped dynamic workflows last week. Claude writes an execution framework for the current task on the spot, and workflows can be saved, shared and reused. They coordinate subagents through agent(), parallel() and pipeline() in JavaScript, can pick a model and a separate worktree, and resume a session after an interruption.

Why it matters: Claude Code splits execution and adversarial verification across subagents with separate contexts, a concrete orchestration method for cutting self-preference and goal drift on complex tasks.

May 26Tuesday

Alibaba Technology · WeChat

Nearly 9x training speedup: residual streams in DiT are becoming a convergence bottleneck

Nanjing University LAMDA and Alibaba Intelligent Engine proposed DAR, a timestep-aware cross-layer routing method that replaces fixed residual accumulation in DiT; on ImageNet 256x256, it reduced SiT-XL/2 FID from 9.67 to 7.56 and reached baseline convergence quality with 8.75x fewer training iterations.

Why it matters: Nanjing University's LAMDA lab and Alibaba replace DiT's fixed residual accumulation with per-timestep cross-layer routing, reaching equal convergence on ImageNet in about one-ninth the training steps.