AI 动态

Building advertising for the way people use AI

OpenAI introduces a new visual ad format in ChatGPT and expands measurement tools, attribution partnerships, and brand suitability for advertisers.

中文翻译待同步，当前显示官方原文。

站内阅读

A model guide for the GPT-6 family

Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.

中文翻译待同步，当前显示官方原文。

站内阅读

Chatham scales its capital markets expertise with OpenAI

Chatham Financial uses Codex and GPT-5.6 to build technology and redesign workflows, cutting trade validation from 30 minutes to under 4.

中文翻译待同步，当前显示官方原文。

站内阅读

How Albertsons Companies is reimagining retail from the inside out

Albertsons Cos. is using ChatGPT Enterprise and the OpenAI API to help teams work faster and make grocery shopping easier for millions of customers.

中文翻译待同步，当前显示官方原文。

站内阅读

The Den frees up 10-15 hours a week to grow with ChatGPT Work

As it opens a new location, the social club prepares grant applications in 2 hours instead of 3 days and liquor-license materials in 3 hours instead of 4 days.

中文翻译待同步，当前显示官方原文。

站内阅读

We've added a line field to the Models API.

We've added a line field to the Models API. GET /v1/models and GET /v1/models/{model_id} now return the model line each model belongs to. Claude Opus 4.5 and Claude Opus 4.6 both report opus, for example. Use line to group models without parsing their IDs. line is null for a model that belongs to no line. See Using the Models API.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini 4 Argon: our next era of frontier intelligence

中文翻译待同步，当前显示官方原文。

站内阅读

Disrupting a coordinated model-distillation campaign

Learn how OpenAI disrupted a campaign to extract protected model reasoning and is strengthening defenses against adversarial distillation.

中文翻译待同步，当前显示官方原文。

站内阅读

We announced the deprecation of the Claude Sonnet 4.5 model (claude-sonnet-4-5-20250929), with retirement on the Claude API scheduled for November 30, 2026.

We announced the deprecation of the Claude Sonnet 4.5 model (claude-sonnet-4-5-20250929), with retirement on the Claude API scheduled for November 30, 2026. We recommend migrating to Claude Sonnet 5.5. Read more in Model deprecations.

中文翻译待同步，当前显示官方原文。

站内阅读

Let skills in Gemini tackle your most repetitive tasks What: Today we’re rolling out skills directly into the Gemini chat globally1.

Let skills in Gemini tackle your most repetitive tasks What: Today we’re rolling out skills directly into the Gemini chat globally1. Skills are a reusable set of instructions that tell Gemini how to handle specific tasks. Think of them as saved shortcuts to faster prompting. Gemini can now help you build custom skills from your chats and automatically run them whenever you enter a matching prompt. For bigger tasks, you can stack multiple skills together at the same time — like combining a writing style skill with a brand guidelines skill to create on-brand assets in your voice. Skills will soon replace Gems as the tool for tailoring instructions for specific tasks. To make this transition seamless, we will automatically migrate your Gems into skills when Gems go away. You can learn more about how to transition your Gems into skills by visiting our help center. Why: We want to help you spend less time prompting and more time doing. With skills, you can save your most-used instructions once and run them endlessly – so you can skip the repetitive prompting and jump straight to the results. 1 currently available to 18+ and coming to more users soon.

中文翻译待同步，当前显示官方原文。

站内阅读

DevDay 2026 Recap

Explore more than 20 announcements from OpenAI DevDay 2026, including GPT-6 Astra, ChatGPT, Codex, APIs, security, and new tools for builders.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing GPT-6.1 Sol

Meet GPT-6.1 Sol: near-Astra intelligence for coding, computer use, and professional work at one-fifth of Astra’s standard API input and output token prices.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing dots

Dots by OpenAI are proactive assistants that can keep working across complex projects and everyday tasks. Learn how dots help you stay in control while work moves forward.

中文翻译待同步，当前显示官方原文。

站内阅读

Basis completes a tax workbook 2x faster with GPT-6 Astra

GPT-6 Astra completed a 50-tab tax workbook twice as fast as GPT-5.6 Sol, and its stronger understanding of user intent gives Basis more confidence in real-world use.

中文翻译待同步，当前显示官方原文。

站内阅读

Are you a Codex Original?

We’re collecting real stories of builders, tinkerers, researchers, and creators who are using Codex to do incredible things. If you want to be a part of the next chapter of the Codex Originals program, tell us more about your story and project below.

中文翻译待同步，当前显示官方原文。

站内阅读

Claude Sonnet 5.5 launch We just launched Claude Sonnet 5.5, the second model in our Claude 5.5 family.

Claude Sonnet 5.5 launch We just launched Claude Sonnet 5.5, the second model in our Claude 5.5 family. Sonnet 5.5 is a faster, lower-cost complement to Claude Opus 5.5. For more information, see our blog post: Introducing Claude Sonnet 5.5.

中文翻译待同步，当前显示官方原文。

站内阅读

We've launched Claude Sonnet 5.5 (claude-sonnet-5-5).

We've launched Claude Sonnet 5.5 (claude-sonnet-5-5). It's available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude on Google Cloud, and Claude in Microsoft Foundry. For its context window, output limits, and prices, see the Claude Sonnet 5.5 model page. Code written for Claude Sonnet 5 can break on Claude Sonnet 5.5 in five ways. To turn off up-front thinking, send thinking: {"type": "between_tools"} instead of "disabled", at high effort or below. Forced tool use (tool_choice types any and tool) returns a 400 error. Thinking blocks are tied to the model and the conversation. On the Claude API and Google Cloud, the earlier computer_20251124 computer use tool isn't accepted. The advisor tool rejects Claude Opus 4.8, Claude Opus 4.7, and Claude Sonnet 5 as advisors. See What's new in Claude Sonnet 5.5 for each change and the migration guide for before-and-after requests. For model-specific prompting patterns, see Prompting Claude Sonnet 5.5. Thinking blocks that Claude Sonnet 5.5 produces work only in the account that produced them, or in an account linked to it. When another account sends one of these blocks, the API drops the block before the model sees it, and the request succeeds. Blocks from earlier models aren't affected. See Preserved thinking.

中文翻译待同步，当前显示官方原文。

站内阅读

Proaction boosts sales 60% and saves 75+ hours with Codex

With Codex, GPT-Live-1, and GPT-6 Astra, Proaction builds, operates, and sells modern fleet management faster.

中文翻译待同步，当前显示官方原文。

站内阅读

Wayfair boosts catalog accuracy and support speed with OpenAI

Wayfair uses OpenAI models to improve ecommerce support and product catalog accuracy, automating ticket triage and enhancing millions of product attributes at scale.

中文翻译待同步，当前显示官方原文。

站内阅读

Build plugins for Claude You can now submit plugins to the Claude directory through a new developer portal, track them through review, and see usage analytics once they’re live.

Build plugins for Claude You can now submit plugins to the Claude directory through a new developer portal, track them through review, and see usage analytics once they’re live. For more information, see our blog post: Build plugins for Claude.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing Gemini 3.8 Live with Live Avatar

中文翻译待同步，当前显示官方原文。

站内阅读

We're resuming billing for refusals that arrive before any output when stop_details.category is "bio", "frontier_llm", or "reasoning_extraction", the categories where we measure lo

We're resuming billing for refusals that arrive before any output when stop_details.category is "bio", "frontier_llm", or "reasoning_extraction", the categories where we measure low volumes of false positives. Mid-stream refusals were already billed. Refusals billed under this change are charged like any other request, at the rates of the model that ran it. Refusals before any output in other categories are still not billed, and fallback credit is unchanged. This change applies on all platforms. See How refusals are billed. The Compliance API local session endpoints are out of beta for Claude for Microsoft 365 sessions in Excel, PowerPoint, Word, and Outlook (product_surface values beginning with office_agents). See Sessions on users' machines. The Compliance API Activity Feed no longer returns file names, project document names, or artifact titles. The filename and title fields on file, project document, and artifact activities are now always empty or omitted, including on activities recorded before this change. To look up a name or title by the ID on the activity, use a Compliance Access Key with the read:compliance_user_data scope. See Understand the Activity object.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini 3.8 text-to-speech says hello

中文翻译待同步，当前显示官方原文。

站内阅读

Ringg’s AI agents resolve up to 65% of customer calls with OpenAI

Using GPT-5.6, Ringg powers multilingual agents across voice, chat, WhatsApp, and web for 90% less cost vs. GPT-4.1.

中文翻译待同步，当前显示官方原文。

站内阅读

Harvey turns legal context into stronger drafts with GPT-6 Astra

GPT-6 Astra produces more structured, context-aware legal documents, freeing lawyers to focus on strategy.

中文翻译待同步，当前显示官方原文。

站内阅读

How invideo improves color grading 3x with GPT‑6 Astra

With GPT‑6 Astra, invideo plans edits with greater precision, improves color correction and grading threefold, and produces 50 custom effects in one day.

中文翻译待同步，当前显示官方原文。

站内阅读

ChatGPT Ads expands to Southeast Asia and Taiwan

ChatGPT Ads is expanding to Southeast Asia and Taiwan, giving eligible businesses new ways to reach people across more than 60 countries.

中文翻译待同步，当前显示官方原文。

站内阅读

Airbnb widens access to GPT-6 Astra and OpenAI frontier models

Learn how Airbnb is expanding access to GPT-6 Astra and OpenAI frontier models to help engineering teams solve bugs, design systems, and ship faster.

中文翻译待同步，当前显示官方原文。

站内阅读

Cache diagnostics is out of beta on the Claude API and no longer requires the cache-diagnosis-2026-04-07 beta header.

Cache diagnostics is out of beta on the Claude API and no longer requires the cache-diagnosis-2026-04-07 beta header. Include the diagnostics object on a Messages request to opt in; requests that still send the header work as before. Responses from POST /v1/messages now always include the diagnostics field, which is null when the request did not include the diagnostics object.

中文翻译待同步，当前显示官方原文。

站内阅读

Better prompt caching for GPT-6

Learn how GPT-6 improves prompt caching with higher cache hit rates, new diagnostics, explicit breakpoints, and controls that reduce latency and costs.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing GPT-6 Sol and Luna

Meet GPT-6 Sol and Luna, two models that bring frontier intelligence to everyday work with different balances of capability and cost.

中文翻译待同步，当前显示官方原文。

站内阅读

Parallel cut research time and cost in half with GPT‑6 Astra

GPT‑6 Astra allowed Parallel’s agents to research and synthesize labor-market data in half the time and at half the cost vs. prior models.

中文翻译待同步，当前显示官方原文。

站内阅读

Priorities and principles for effective third party assessments

OpenAI outlines priorities and principles for rigorous, secure, and independent third-party AI safety assessments of frontier models and safeguards.

中文翻译待同步，当前显示官方原文。

站内阅读

Claude Opus 5.5 launch We just launched Claude Opus 5.5, the first model in our new Claude 5.5 family.

Claude Opus 5.5 launch We just launched Claude Opus 5.5, the first model in our new Claude 5.5 family. It performs at the level of Claude Fable 5.1 on most work and costs 40% less to run than Opus 5. For more information, see our blog post: Introducing Claude Opus 5.5.

中文翻译待同步，当前显示官方原文。

站内阅读

We've launched Claude Opus 5.5 (claude-opus-5-5), a model for long-running agentic coding and knowledge work.

We've launched Claude Opus 5.5 (claude-opus-5-5), a model for long-running agentic coding and knowledge work. It has a 1M token context window by default, 128k max output tokens, and always-on adaptive thinking, at $4 / $20 USD per MTok (Claude Opus 5 is $5 / $25). Claude Opus 5.5 is available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude on Google Cloud, and Claude in Microsoft Foundry. See What's new in Claude Opus 5.5 for capabilities, API changes, and migration guidance. On Claude Opus 5.5, thinking can't be disabled: thinking: {"type": "disabled"} and thinking: {"type": "enabled", ...} return a 400 error. Omit the thinking field and control thinking depth with the effort parameter. tool_choice types any and tool also return a 400 error, as on Claude Fable 5.1; use auto with strict tool use. On the Claude API and Google Cloud, computer use on this model requires the computer_toolset_20260801 toolset and the earlier computer_20251124 tool returns a 400 error; on Amazon Bedrock, computer_20251124 keeps working. See the migration guide. Fast mode (research preview) is available for Claude Opus 5.5 on the Claude API. Tools can now be defined inside a mid-conversation system message, in beta on the Claude API with the inline-tools-2026-09-15 beta header. A tool_addition block can carry the tool's full definition (tool: {"type": "tool_definition", "definition": {...}}), so you can add a tool, change its schema, or move a server tool to a newer version without editing tools or invalidating the prompt cache. The same header covers adding and removing tools by reference. With the MCP connector's mcp-client-2026-09-15 beta header as well, the definition can be an MCP toolset, and a response records each server's fetched tool list in an mcp_tool_listing block, which pins that list when you send it back.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS generally available (GA): Released our next-generation text-to-speech (TTS) audio models and the Gemini API Voices endpoint (/v1b

Gemini 3.8 Flash TTS and Gemini 3.8 Flash-Lite TTS generally available (GA): Released our next-generation text-to-speech (TTS) audio models and the Gemini API Voices endpoint (/v1beta/voices): Gemini 3.8 Flash TTS (gemini-3.8-flash-tts): Flagship creative TTS model engineered for studio-grade voice fidelity, nuanced acting, regional dialects, and long-form multi-turn stability. Gemini 3.8 Flash-Lite TTS (gemini-3.8-flash-lite-tts): Fast, cost-efficient TTS model built to replace gemini-3.1-flash-tts-preview for high-throughput production and real-time voice agent cascades. Voice design, Voice replication, and the Extended Voice Library: Create persistent custom vocal personas from text prompts, replicate voices with consent verification, and query 150+ prebuilt and custom voices. See the Text-to-speech guide to get started.

中文翻译待同步，当前显示官方原文。

站内阅读

Higgsfield AI ships new video features in a day with GPT-6 Astra

With GPT-6 Astra, Higgsfield AI makes video ad creation easier for small businesses and brings new creative tools to market faster.

中文翻译待同步，当前显示官方原文。

站内阅读

V7 cuts costs 78% while boosting accuracy with GPT-5.6 Luna

Using GPT-5.6, V7 turns scattered company files into context agents can use to complete complex, source-linked work.

中文翻译待同步，当前显示官方原文。

站内阅读

For cache diagnostics, a response to a request that sends the cache-diagnosis-2026-04-07 beta header now always includes the diagnostics field.

For cache diagnostics, a response to a request that sends the cache-diagnosis-2026-04-07 beta header now always includes the diagnostics field. The field is null when the request did not include the diagnostics object. Previously the field was omitted in that case. The Compliance API local session endpoints now also return transcripts of Claude in Chrome sessions (product_surface value claude_in_chrome), in beta for Claude Enterprise organizations, with your existing Compliance Access Key and the read:compliance_user_data scope. See Sessions on users' machines.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini 2.5 models access update: To ensure reliable performance for everyone, we are limiting access to the 2.5 models to users who have actively used them in the past.

Gemini 2.5 models access update: To ensure reliable performance for everyone, we are limiting access to the 2.5 models to users who have actively used them in the past. These models are not deprecated and will continue to be served until further notice through the API. For any new projects, use our latest models: 3.5 Flash-Lite or 3.8 Flash. This helps us maintain sufficient capacity for both ongoing legacy workflows and new applications.

中文翻译待同步，当前显示官方原文。

站内阅读

How Cooley is accelerating IPO work with ChatGPT

Cooley built GO Public with ChatGPT Work to bring intelligence to the IPO process, helping lawyers surface issues earlier and focus judgment where it matters most.

中文翻译待同步，当前显示官方原文。

站内阅读

Antigravity Agent 09-2026: Released antigravity-preview-09-2026, which replaces and deprecates antigravity-preview-05-2026.

Antigravity Agent 09-2026: Released antigravity-preview-09-2026, which replaces and deprecates antigravity-preview-05-2026. If you run on a remote sandbox (environment: "remote") and read only output_text or model_output steps, update the agent string and nothing else changes. If you run tools locally (local_environment) or parse function_call steps, the built-in tools changed. Parameters use PascalCase instead of snake_case, and file edits use line-range replacements instead of full rewrites. Capability 05-2026 09-2026 File creation write_file(path, content) write_to_file(TargetFile, CodeContent, Overwrite, Description) File editing write_file(path, content), full rewrite replace_file_content(TargetFile, StartLine, EndLine, TargetContent, ReplacementContent) File reading read_file(path, offset, limit), byte offsets view_file(AbsolutePath, StartLine, EndLine, ContentOffset) Directory listing list_files(path) list_dir(DirectoryPath) File and code search None, agents used shell commands find_by_name(SearchDirectory, Pattern, MaxDepth) and grep_search(SearchPath, Query, IsRegex) Shell execution code_execution(command, timeout_seconds) Unchanged Web search google_search(queries) Unchanged See the Antigravity Agent guide. antigravity-preview-05-2026 shuts down on October 5, 2026, tracked on the deprecations page.

中文翻译待同步，当前显示官方原文。

站内阅读

Our framework for reporting model misalignment

OpenAI shares a framework for tracking, investigating, and disclosing model misalignment, alongside six reports of unexpected or concerning model behavior.

中文翻译待同步，当前显示官方原文。

站内阅读

Helping older adults use AI in everyday life

OpenAI and AARP are bringing free, hands-on ChatGPT workshops to 1,000 older adults across 10 U.S. cities to build practical AI skills safely.

中文翻译待同步，当前显示官方原文。

站内阅读

Reimagining advertising with AI

Explore new AI-powered advertising experiences from OpenAI, including Sponsored Agents, tools for marketers, and integrations with HubSpot and Shopify.

中文翻译待同步，当前显示官方原文。

站内阅读

Hex turns complex analysis into visual reports with GPT‑6 Astra

GPT-6 Astra helps Hex’s data agents turn answers into interactive visualizations that employees are proud to share.

中文翻译待同步，当前显示官方原文。

站内阅读

How to connect AI usage to business value

Learn how ChatGPT Work and Codex analytics help teams understand AI usage and spend, identify training needs, and connect adoption to business outcomes.

中文翻译待同步，当前显示官方原文。

站内阅读

Claude Cowork comes to every conversation We're rolling out a new Claude experience that makes everything Claude Cowork does available from any conversation.

Claude Cowork comes to every conversation We're rolling out a new Claude experience that makes everything Claude Cowork does available from any conversation. Ask a quick question or hand Claude a bigger task, like a report, spreadsheet, or presentation, without choosing a mode first. Your chats, Cowork tasks, projects, connectors, and skills carry over. The new experience is rolling out gradually to Pro and Max plans on web, desktop, and mobile. Learn more in Claude Cowork and chat are one Claude. Create designs, decks, and docs in any conversation You can now ask for a design, deck, or document in any conversation with Claude, including in Claude Code and the Artifacts tab, and edit it with Claude or directly as you work. Claude Design works inside your conversations with all of its features, including on-canvas editing and importing your design system. Claude Slides gives presentations their own starting point, and Claude Docs is a new way to write living documents with Claude and your team. Artifacts, including Claude Design, Claude Slides, and Claude Docs, are available on every plan, including Free. On Enterprise plans, Claude Design, Claude Slides, and Claude Docs are in beta and off by default until an owner turns them on. Learn more in What are artifacts and how do I use them? and Get started with Claude Docs.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing Gemini 3.8 Live and 3.8 Live Extended Thinking

中文翻译待同步，当前显示官方原文。

站内阅读

Launching Salesforce in Claude (beta) We've launched Salesforce in Claude, a plugin that brings a seller’s accounts, opportunities, and pipeline into Claude with 37 pre-built sales

Launching Salesforce in Claude (beta) We've launched Salesforce in Claude, a plugin that brings a seller’s accounts, opportunities, and pipeline into Claude with 37 pre-built sales skills. Prep a call, review a deal, create a pipeline dashboard, and send your forecast all from Claude. It’s now available in beta on all paid plans for organizations Salesforce approves through its beta sign-up. For more information, see our blog post: Bringing Salesforce into Claude.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking generally available (GA): Released two new audio-to-audio models for real-time voice applications using the Live API: Gemini 3

Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking generally available (GA): Released two new audio-to-audio models for real-time voice applications using the Live API: Gemini 3.8 Live (gemini-3.8-live): The default option for most low-latency voice agent experiences and real-time dialogue without reasoning delays. Features interleaved reasoning, default asynchronous function calling, and full session client content updates. Gemini 3.8 Live Extended Thinking (gemini-3.8-live-extended-thinking): High-reasoning audio-to-audio model supporting background reasoning during live audio interactions, recommended when higher background reasoning is required. To get started, see the Live API guide, the Capabilities guide, and the Thinking guide.

中文翻译待同步，当前显示官方原文。

站内阅读

How Fyxer built an AI executive assistant people trust

Fyxer uses OpenAI models, fine-tuning, memory, and real user feedback to organize inboxes and draft emails in each user’s voice.

中文翻译待同步，当前显示官方原文。

站内阅读

Perplexity trusts GPT-6 Astra with end-to-end systems

Perplexity uses Astra to write communications, change software, and monitor production systems, and checks in much less frequently than with earlier models.

中文翻译待同步，当前显示官方原文。

站内阅读

The Messages API can now compact a conversation on demand on the Claude API, in beta with the compact-2026-09-04 beta header.

The Messages API can now compact a conversation on demand on the Claude API, in beta with the compact-2026-09-04 beta header. Send the top-level compaction parameter, and the API returns a signed compaction block that summarizes the messages you sent. On later requests, send that block first, in place of those messages. You choose when to compact, the request can run in the background, and you can keep recent turns word for word after the summary. On models with preserved thinking, the thinking in those kept turns can stay valid.

中文翻译待同步，当前显示官方原文。

站内阅读

Cognition helps Devin test its own work with GPT‑6 Astra

GPT‑6 Astra improves Devin’s ability to test software and show that it works, with the goal of helping engineers review less code and ship more.

中文翻译待同步，当前显示官方原文。

站内阅读

Rapidly scaling online storage to serve over 1 billion ChatGPT users

Learn how OpenAI evolved Habitat from a Python library into a globally distributed storage platform serving 1 billion ChatGPT users and 22M requests per second.

中文翻译待同步，当前显示官方原文。

站内阅读

How a researcher uses Codex and ChatGPT to search for new antimicrobial molecules

César de la Fuente’s lab uses Codex and ChatGPT to search living and extinct genomes for antimicrobial candidates to fight drug-resistant infections.

中文翻译待同步，当前显示官方原文。

站内阅读

Now everyone can put data to work

Meet the Data agent in ChatGPT Work. Connect company data, uncover insights, and build interactive dashboards with AI using natural language.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing ChatGPT for Financial Services

Introducing ChatGPT for Financial Services, combining built-in financial data and GPT-6 Astra for research, modeling, and client-ready materials.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing the Agents API

Build and launch cloud agents with the Agents API, a managed service powered by the Codex harness for orchestration, long-running sessions, and tool use.

中文翻译待同步，当前显示官方原文。

站内阅读

Build more natural voice experiences with GPT‑Live‑1 in the API

GPT‑Live‑1 brings natural, full-duplex voice conversations to the API, with stronger instruction following, custom voices, and telephony support.

中文翻译待同步，当前显示官方原文。

站内阅读

Smart reports (beta) We’ve launched smart reports, which analyze how a team uses Claude and report on the work getting done, what it costs, where sessions run into friction, and wh

Smart reports (beta) We’ve launched smart reports, which analyze how a team uses Claude and report on the work getting done, what it costs, where sessions run into friction, and which repeated patterns are worth packaging as shared skills. Smart reports are available in beta on Claude Enterprise plans. Learn more in Get started with smart reports.

中文翻译待同步，当前显示官方原文。

站内阅读

Claude Managed Agents permission policies now include auto: the server evaluates each agent or MCP tool call and runs it, denies it, or pauses for your approval.

Claude Managed Agents permission policies now include auto: the server evaluates each agent or MCP tool call and runs it, denies it, or pauses for your approval. agent.tool_use and agent.mcp_tool_use events report how each call was evaluated in an evaluation field alongside evaluated_permission. See Let the server evaluate each call with auto. The ant CLI adds ant beta:sessions connect, which attaches your terminal to a Claude Managed Agents session. You can follow the session live, send messages, and allow or deny tool calls that are waiting for approval. Pass --web to serve the Claude Console's session viewer locally and open the session there instead. See Connect to a Managed Agents session from your terminal.

中文翻译待同步，当前显示官方原文。

站内阅读

Bring Google Gemini to your PC with the new Windows app What: The Gemini app is now available on Windows.

Bring Google Gemini to your PC with the new Windows app What: The Gemini app is now available on Windows. Press Alt + Space to summon Gemini instantly from anywhere on your PC whenever you need it. You can get quick answers, draft messages, and brainstorm ideas right alongside your active applications. Inside your dedicated desktop workspace, you have access to everything you love about Gemini. You can connect to Google apps like Gmail and Google Drive to summarize documents and find information quickly, generate custom images with Nano Banana, and dive into in-depth research on any topic. The app is available on Windows 10 or later. Download it today at gemini.google/desktop. Why: Your desktop is where you get work done. Bringing Gemini straight to your PC with instant shortcut access keeps you in your flow, making it simple to get answers and create visuals right alongside your everyday tools.

中文翻译待同步，当前显示官方原文。

站内阅读

DeepSeek-V4.1-Flash 发布​

DeepSeek-V4.1-Flash 发布​ 今天，我们正式发布 DeepSeek-V4.1-Flash 模型。这是我们全新模型结构系列中的最小尺寸的模型，具备原生多模态视觉理解能力。新模型结构的设计初衷是：能力上限更高、推理速度更快、吞吐更大、可扩展到更大参数模型。 GPQA Diamond: 90.9 HLE: 36.8 (39.1*) Codeforces (Rating): 3471 MathArena Apex: 65.6 Terminal-Bench 2.1: 90.6 Terminal-Bench 3.0: 30.0 Terminal-Bench 4.0: 31.2 DeepSWE v1.1: 74.2 ProgramBench: 20.3 NL2Repo-Bench: 65.4 CyberGym: 88.1 SEC-Bench Pro: 62.8 ExploitGym: 15.3 HLE (w/tools): 63.9 Automation-Bench: 54.8 Agents' Last Exam: 31.8 Chartography (w/tools): 78.9 BabyVision (w/tools): 89.6 ZeroBench-main (w/tools): 49.0 * 表示仅在 HLE 评测集的纯文本子集上进行了测试。 API 变更 DeepSeek V4.1 Flash 已同步上线 DeepSeek API，原生支持多模态，将模型名称更改为 deepseek-flash 即可调用最新的 V4.1 Flash 模型。旧版本模型 V4 Flash 与 V4 Flash Vision Exp 现已下线，出于兼容考虑，模型名 deepseek-v4-flash、deepseek-v4-flash-vision-exp 将被暂时路由到 V4.1 Flash。 为响应广大用户的需求，我们决定在 2026 年 9 月 14 日之后继续提供 DeepSeek V4 Pro 的 API 调用服务，计费方式保持不变；如有变动，我们将另行通知。感谢您的理解与支持！ API 定价调整 随着 DeepSeek-V4.1-Flash 上线，API 价格同步下调。详细价格请参考模型与价格。 详细更新内容请参阅文档。

站内阅读

GPT-6 Astra: The next generation in intelligence for work

Meet GPT-6 Astra, OpenAI’s most capable model for business, with advanced reasoning, computer use, and stronger writing and design judgment.

中文翻译待同步，当前显示官方原文。

站内阅读

For cache diagnostics, the API now stores a request's fingerprint only when the request includes the diagnostics object.

For cache diagnostics, the API now stores a request's fingerprint only when the request includes the diagnostics object. A request that sends only the cache-diagnosis-2026-04-07 beta header is still accepted, but no fingerprint is stored. A later turn that points previous_message_id at it reports previous_message_not_found. Include diagnostics on every turn, with "previous_message_id": null on the first.

中文翻译待同步，当前显示官方原文。

站内阅读

How GPT-5.6 Sol helps run quantum computing experiments

See how an MIT researcher uses GPT-5.6 Sol with Codex to autonomously run quantum computing experiments, analyze results, and calibrate qubits.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing ChatGPT Images 2.5

ChatGPT Images 2.5 helps turn your ideas, sketches, and reference photos into more personalized, polished images that better reflect your ideas.

中文翻译待同步，当前显示官方原文。

站内阅读

On the Navier–Stokes Millennium Prize Problem

We’re sharing an AI-generated solution to the Navier–Stokes Millennium Prize Problem, including a writeup and a formal proof in Lean.

中文翻译待同步，当前显示官方原文。

站内阅读

1Password increases engineering productivity 21% with Codex

Engineers at 1Password use Codex to rapidly build new features and internal tools, reaching production-readiness while maintaining rigorous security policies.

中文翻译待同步，当前显示官方原文。

站内阅读

Research acceleration: The view inside OpenAI

Inside OpenAI, coding agents are reshaping AI research. Explore early data on agent usage, experiment velocity, task complexity, and research acceleration.

中文翻译待同步，当前显示官方原文。

站内阅读

Legora reviewed 41 documents in minutes with GPT-6 Astra

Legora used GPT-6 Astra to review 41 documents in minutes, find all four planted errors, and improve performance by nearly 40% in this financial-review workflow.

中文翻译待同步，当前显示官方原文。

站内阅读

Playco cut manual fixes 50% prototyping games with GPT-6 Astra

Using GPT-6 Astra, Playco built three themed game prototypes from one grey box foundation and reported 50% fewer manual fixes than with the previous model.

中文翻译待同步，当前显示官方原文。

站内阅读

GPT-6 Astra: A new generation of intelligence

Introducing GPT-6 Astra, our most intelligent and aligned model yet, with state-of-the-art capabilities across computer use, coding, cybersecurity, and science.

中文翻译待同步，当前显示官方原文。

站内阅读

Safety overview: GPT-6 Astra

GPT-6 Astra is our most capable broadly deployed model and our first to reach the Critical level of cybersecurity capability under our Preparedness Framework.

中文翻译待同步，当前显示官方原文。

站内阅读

Version 1.30.0 of the ant CLI adds ant apply, which creates and updates agents, environments, skills, memory stores, and deployments from files in your repository.

Version 1.30.0 of the ant CLI adds ant apply, which creates and updates agents, environments, skills, memory stores, and deployments from files in your repository. Describe each resource in a file, run ant apply, and approve the plan it prints. Commit the claude-lock.json lockfile it writes so that later runs, on your machine or in CI, update the same resources instead of creating new ones. See Manage resources as code with ant apply.

中文翻译待同步，当前显示官方原文。

站内阅读

Lyria 3.5 generally available (GA): Released the next generation of Google's music generation model: lyria-3.5: Full-length song generation with improved musical coherence, natural

Lyria 3.5 generally available (GA): Released the next generation of Google's music generation model: lyria-3.5: Full-length song generation with improved musical coherence, natural vocals, and fine-grained duration and structural control. The model supports text and image inputs and generates high-fidelity 44.1 kHz stereo audio. See the Music generation guide for details and code samples.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing Gemini 3.8 Flash and 3.8 Flash Cyber

中文翻译待同步，当前显示官方原文。

站内阅读

ATV Big Air Tour turned 3 days of work into 3 hours with ChatGPT

ATV Big Air Tour uses ChatGPT Work to speed up marketing, merchandising, and more. It even turned merchandise photos into an inventory website in 15 minutes.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini 3.8 Flash generally available (GA): Released gemini-3.8-flash, our most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and com

Gemini 3.8 Flash generally available (GA): Released gemini-3.8-flash, our most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows. To get started, see the Gemini 3.8 Flash model page and the Latest model guide.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing agentic video understanding with Gemini

中文翻译待同步，当前显示官方原文。

站内阅读

Path to Astra: critical capabilities and frontier safeguards

Astra is the first OpenAI model to meet the Critical cybersecurity capability threshold under the Preparedness Framework, with stronger safeguards for release.

中文翻译待同步，当前显示官方原文。

站内阅读

Healthcare organizations can now connect EHR and additional industry data to ChatGPT

ChatGPT can now connect to trusted healthcare data, helping clinicians securely access patient context, medical research, and more.

中文翻译待同步，当前显示官方原文。

站内阅读

How law firm Gilbert + Tobin governs and scales AI with OpenAI

See how Gilbert + Tobin combines CEO-led commitment, rigorous governance, and human accountability to scale ChatGPT Enterprise and Codex across the firm.

中文翻译待同步，当前显示官方原文。

站内阅读

Claude Fable 5.1 and Claude Mythos 5.1 launch We just launched Claude Fable 5.1 and Claude Mythos 5.1, the world’s most advanced models for coding and knowledge work.

Claude Fable 5.1 and Claude Mythos 5.1 launch We just launched Claude Fable 5.1 and Claude Mythos 5.1, the world’s most advanced models for coding and knowledge work. For more information, see our blog post: Claude Fable 5.1 and Mythos 5.1.

中文翻译待同步，当前显示官方原文。

站内阅读

We've launched Claude Fable 5.1 (claude-fable-5-1), the successor to Claude Fable 5 for long-running agentic coding, knowledge work, and research, alongside Claude Mythos 5.1 (clau

We've launched Claude Fable 5.1 (claude-fable-5-1), the successor to Claude Fable 5 for long-running agentic coding, knowledge work, and research, alongside Claude Mythos 5.1 (claude-mythos-5-1) for Project Glasswing participants. Both models support a 1M token context window by default, 128k max output tokens, and always-on adaptive thinking, at $10 / $50 USD per MTok, the same as Claude Fable 5, with cache reads cut to $0.25 per MTok. Claude Fable 5.1 is available on the Claude API, Claude in Amazon Bedrock, Claude Platform on AWS, Claude on Google Cloud, and Claude in Microsoft Foundry. See What's new in Claude Fable 5.1 for capabilities, API changes, and migration guidance. Prompt cache reads on Claude Fable 5.1 and Claude Mythos 5.1 cost $0.25 USD per million tokens: 0.025x the base input price, compared with 0.1x on other models. Cache writes are unchanged. See Prompt caching pricing. On Claude Fable 5.1 and Claude Mythos 5.1, tool_choice types any and tool aren't supported and return a 400 error. auto and none are unchanged. To guarantee schema-conformant tool inputs, use strict tool use or structured outputs. Thinking blocks produced by Claude Fable 5.1 and Claude Mythos 5.1 are preserved only for the model that produced them or a newer one: earlier models can't read them, and the API drops one replayed to an earlier model. Claude Fable 5.1 accepts thinking blocks from Claude Opus 5, Claude Fable 5, Claude Mythos 5, and earlier Claude models. On Claude Fable 5.1, the API also checks that nothing before a block has changed: for new accounts created on or after August 31, 2026, replaying one after the system prompt, tools, or an earlier message changed returns a 400 error. With the thinking-binding-controls-2026-08-01 beta header, dropped blocks are reported in an input_transformations response field, and thinking.block_binding.prefix_mismatch_behavior chooses between rejecting and dropping blocks whose history changed. See Preserved thinking. Per-message effort changes are in beta on Claude Fable 5.1, Claude Mythos 5.1, and Claude Opus 5 on the Claude API. Add a role: "system" message with output_config.effort inside messages to change effort for later turns while preserving the prompt cache. Include the mid-conversation-output-config-2026-07-01 beta header in your requests. See Per-message effort. Turn-scoped system messages are in beta (mid-conversation-system-clear-at-2026-08-21 header). Set clear_at: "next_user_message" on a mid-conversation role: "system" message and it renders for the current turn only, then stays in the history at no token cost. Per-turn reminders don't accumulate and don't invalidate the prompt cache or later thinking blocks. thinking.display accepts a third value, "updates", in beta (thinking-display-updates-2026-08-18 header). Reasoning comes back with an empty thinking field, as under "omitted", and the short progress updates that Claude Fable 5.1, Claude Mythos 5.1, and Claude Fable 5 write between tool calls come back as text, at most one thinking block before a tool call. See Progress updates between tool calls. Text generated by Claude Fable 5.1 and Claude Mythos 5.1 carries Anthropic's text watermark, and supported image and video files that Claude produces through the code execution tool carry C2PA Content Credentials when you retrieve them through the Files API on the Claude API. Marking requires no changes to your requests or response handling. Like Claude Fable 5, both models require 30-day data retention and aren't available under zero data retention unless expressly authorized by Anthropic. See Model-specific data retention requirements.

中文翻译待同步，当前显示官方原文。

站内阅读

Agentic video understanding: Released agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite across the Interactions and GenerateContent APIs.

Agentic video understanding: Released agentic video understanding for Gemini 3.7 Flash, 3.6 Flash, and 3.5 Flash-Lite across the Interactions and GenerateContent APIs. The model dynamically navigates video timelines, requesting transcripts, frames, or audio tracks on demand. This approach uses up to 88% fewer tokens for long-form content compared to static processing. To get started, see the Agentic video understanding guide.

中文翻译待同步，当前显示官方原文。

站内阅读

Polimill builds Japan's next-generation public AI infrastructure

Polimill uses OpenAI GPT models and Codex to help municipalities search and use administrative knowledge while accelerating development.

中文翻译待同步，当前显示官方原文。

站内阅读

A milestone in expanding access to AI

ChatGPT Ads reaches $1 billion in annualized revenue run rate and expands globally, supporting broader access to AI through free and affordable options.

中文翻译待同步，当前显示官方原文。

站内阅读

Our decision on Cursor following its acquisition by SpaceX

Our decision to wind down our contract providing OpenAI models to Cursor following its acquisition by SpaceX.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini Omni 1.1 Flash lets you build with more control

中文翻译待同步，当前显示官方原文。

站内阅读

Better answers, broader thinking: What students gain from ChatGPT and critical-thinking training

A randomized study of more than 1,000 students examines ChatGPT, critical thinking, originality, and student performance on a real-world university assignment.

中文翻译待同步，当前显示官方原文。

站内阅读

In Python SDK 1.2.0, TypeScript SDK 0.122.0, Go SDK 1.68.0, Java SDK 2.59.0, Ruby SDK 1.67.0, and C# SDK 12.44.0, client.beta.files and client.beta.skills no longer send the files-

In Python SDK 1.2.0, TypeScript SDK 0.122.0, Go SDK 1.68.0, Java SDK 2.59.0, Ruby SDK 1.67.0, and C# SDK 12.44.0, client.beta.files and client.beta.skills no longer send the files-api-2025-04-14 and skills-2025-10-02 beta headers and return the same shapes as client.files and client.skills. With this change, client.beta.skills.delete() deletes a Skill together with all of its versions, and the beta Messages type BetaSkill (the container Skill reference) is renamed BetaContainerSkill. Requests that still send the beta headers keep receiving the beta shapes. See Migrate from files-api-2025-04-14 and Migrate from skills-2025-10-02. You can now create personal keys and service account keys in the Claude Console. They act as you or as a service account, with the same permissions, and stop working when the linked account is removed from an organization. This lets organization admins more easily track usage for each account, and ensure key usage is legitimate. These API keys can be scoped to a specific workspace or work on admin endpoints and across any workspace the account has access to. Workspace API keys remain supported as a legacy option. See API keys for more information.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini Omni Flash generally available (GA): Released gemini-omni-1.1-flash, the GA version of our fast, conversational video generation and editing model.

Gemini Omni Flash generally available (GA): Released gemini-omni-1.1-flash, the GA version of our fast, conversational video generation and editing model. This release includes significant new capabilities: Video extension: Seamlessly extend existing videos by generating continuations at the end of a clip using the extend task or directly with a prompt. Interpolation (first + last frame): Generate a video transitioning between two images using the image_to_video task with up to 2 images. Resolution control: New resolution parameter in video_config supports 360p, 720p (default), 1080p, and 4k outputs. 1080p and 4K outputs are generated using upscaling. The existing gemini-omni-flash-preview endpoint will be deprecated on September 30, 2026. To get started, see the Gemini Omni Flash model page and the omni guide.

中文翻译待同步，当前显示官方原文。

站内阅读

Intelligent transcription with Gemini 3.5 Transcribe

Now you can get more intelligent speech-to-text transcription with Gemini 3.5 Transcribe.

中文翻译待同步，当前显示官方原文。

站内阅读

Bringing ChatGPT for Teachers to more U.S. school districts

ChatGPT for Teachers is expanding to 55 U.S. school systems, bringing secure AI tools, training, and support to over 100,000 more educators and staff.

中文翻译待同步，当前显示官方原文。

站内阅读

Learning never stops: How AI makes learning continuous

OpenAI’s new report explores how students and educators use ChatGPT to make learning more continuous, with support that extends beyond the classroom.

中文翻译待同步，当前显示官方原文。

站内阅读

The Hugging Face incident and the road ahead

OpenAI shares findings from the Hugging Face security incident and the steps we’re taking to strengthen AI model security, monitoring, and alignment.

中文翻译待同步，当前显示官方原文。

站内阅读

How loveholidays is making everyone a builder with Codex

Discover how loveholidays uses OpenAI Codex to make software development accessible across the business, helping teams turn ideas into products faster.

中文翻译待同步，当前显示官方原文。

站内阅读

The Compliance API session endpoints are out of beta for Cowork and Claude Code sessions.

The Compliance API session endpoints are out of beta for Cowork and Claude Code sessions. See Retrieve session transcripts. The Compliance API local session endpoints now also return transcripts of Claude Science sessions (product_surface value claude_science) and Claude for Microsoft 365 sessions in Excel, PowerPoint, Word, and Outlook (product_surface values beginning with office_agents), in beta for Claude Enterprise organizations, with your existing Compliance Access Key and the read:compliance_user_data scope. See Sessions on users' machines. The Admin API is now available in the ant CLI and the Python, TypeScript, C#, Go, Java, PHP, and Ruby SDKs under client.beta.organization. They cover organization info, members, invites, workspaces and workspace members, API keys, rate limits, service accounts, workload identity federation issuers and rules, and customer-managed encryption keys. Usage and cost reports and the Claude Enterprise user-management and analytics endpoints remain curl-only. The CLI and SDKs read an Admin API key from ANTHROPIC_API_KEY or an org:admin OAuth token from ANTHROPIC_AUTH_TOKEN.

中文翻译待同步，当前显示官方原文。

站内阅读

Gemini 3.5 Transcribe generally available (GA): Released two dedicated speech-to-text models based on Gemini's audio understanding: Gemini 3.5 Transcribe (gemini-3.5-transcribe): H

Gemini 3.5 Transcribe generally available (GA): Released two dedicated speech-to-text models based on Gemini's audio understanding: Gemini 3.5 Transcribe (gemini-3.5-transcribe): High-accuracy, low-latency non-streaming speech-to-text with utterance-based language detection across 85+ languages, speaker diarization, word-level timestamps, and custom vocabulary biasing (up to 1,000 terms). Gemini 3.5 Transcribe Live (gemini-3.5-transcribe-live): Low-latency, bidirectional streaming speech-to-text over WebSockets using the Live API, supporting interim and finalized transcription events, Smart transcription mode, and multiple Voice Activity Detection (VAD) strategies. To get started, see the Audio transcription guide, the Live transcription guide, and the Gemini 3.5 Transcribe model page.

中文翻译待同步，当前显示官方原文。

站内阅读

The full stack behind abundant intelligence

OpenAI CFO Sarah Friar explains how advances across chips, compute, models, and products compound to deliver more useful intelligence at greater scale and lower cost.

中文翻译待同步，当前显示官方原文。

站内阅读

Jalapeño’s first results show industry-leading speed and efficiency in AI inference

Jalapeño is a custom inference chip from OpenAI that delivers faster, more power-efficient AI inference, with higher throughput and lower latency for modern models.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing the Admin plugin for ChatGPT Work and Codex

Use the Admin plugin for ChatGPT Work and Codex to analyze workspace usage, manage members and permissions, adjust limits, and act on admin requests.

中文翻译待同步，当前显示官方原文。

站内阅读

Memory in Claude Cowork, editable topics, and a sensitive topics setting Memory now works across chat and Cowork in the cloud.

Memory in Claude Cowork, editable topics, and a sensitive topics setting Memory now works across chat and Cowork in the cloud. Everything Claude remembers is listed under Topics in Settings > Memory, where you can edit or delete any item. Topics such as health or beliefs stay out of memory unless you turn on Include sensitive topics in memory. Memory is on by default for Free, Pro, and Max plans and off by default for Team and Enterprise organizations. See Use Claude's chat search and memory to build on previous context.

中文翻译待同步，当前显示官方原文。

站内阅读

Advancing price-performance for developers with GPT‑5.6 in Kiro

GPT‑5.6 is now available in Kiro, helping developers plan, build, review, and test software with better price-performance.

中文翻译待同步，当前显示官方原文。

站内阅读

DeepSeek-V4-Flash-Vision-Exp 发布​

DeepSeek-V4-Flash-Vision-Exp 发布​ 今天，全新的多模态视觉理解模型 DeepSeek-V4-Flash-Vision-Exp 上线 DeepSeek API 平台，这是一个实验性质的模型，用户可以通过设置 model='deepseek-v4-flash-vision-exp' 访问该模型。 Terminal Bench 2.1: 83.9 NL2Repo: 57.7 DeepSWE: 59.3 DSBench-Hard: 63.6 AutomationBench (Public): 25.7 ApexBench (Pass@1): 36.5 Agents' Last Exam: 27.3 Chartography: 64.3 ZeroBench (Pass@5): 35.0 * 对于公开基准测试集中的 Code Agent 文本任务，DeepSeek 系列模型使用 DeepSeek Harness 极简模式作为框架进行测试，并使用 max 档位，topp=0.95，temperature=1.0；在 ApexBench 与 Agents' Last Exam 测评中，文本模型 DeepSeek-V4-Flash 会忽略其中的多模态元素 在纯文本能力（Agent、推理、世界知识等）方面，DeepSeek-V4-Flash-Vision-Exp 与 DeepSeek-V4-Flash 正式版持平。 在需要视觉理解的 Agent Benchmark 上，DeepSeek-V4-Flash-Vision-Exp 相比 DeepSeek-V4-Flash 实现了大幅跃升，多模态 Agent 能力已接近 Opus-4.8。 详细调用方法请参阅文档。

站内阅读

Stampli cuts launch hours by 68% using ChatGPT Work

With a fixed deadline and design resources committed elsewhere, Stampli used Codex and ChatGPT Work to compress weeks of launch production into days.

中文翻译待同步，当前显示官方原文。

站内阅读

We've released v1.0 of the Python SDK.

We've released v1.0 of the Python SDK. The SDK's HTTP layer moves from httpx to httpx2, a maintained, API-compatible fork: build custom http_client, Timeout, and transport objects from httpx2 (the DefaultHttpxClient helpers are unchanged), and call httpx2.alias_httpx() at startup if you rely on tracing or mocking libraries that patch httpx. v1.0 requires Python 3.10 or later and removes long-deprecated surface, including the legacy Text Completions API, the temperature, top_p, and top_k parameters on Messages methods, and the tool runner's client-side compaction_control. On the async client, .with_raw_response results now need await response.parse(), and AnthropicBedrock now raises an error when no AWS region is configured instead of defaulting to us-east-1. See the v1 migration guide for every change with before-and-after snippets.

中文翻译待同步，当前显示官方原文。

站内阅读

Offering Zero Data Retention for frontier models

OpenAI reaffirms Zero Data Retention for eligible API customers and previews Private Safety Processing for advanced AI safety without compromising data privacy.

中文翻译待同步，当前显示官方原文。

站内阅读

Replit expands access to software creation with GPT-5.6 Luna

Replit introduces Free Mode, powered by GPT-5.6 Luna, so anyone can turn ideas into working software without worrying about token costs.

中文翻译待同步，当前显示官方原文。

站内阅读

The computer use tool is out of beta on the Claude API as the computer_toolset_20260801 toolset: no beta header, batch actions (several actions in one turn), zoom enabled by defaul

The computer use tool is out of beta on the Claude API as the computer_toolset_20260801 toolset: no beta header, batch actions (several actions in one turn), zoom enabled by default, and per-member configuration through configs. Earlier beta versions remain available. Upgrading an existing integration changes the request shape and tool handling; see Migrate from computer_20251124. We've launched the browser use tool (browser_toolset_20260801), a client toolset for driving a browser that your application hosts. It works inside a browser viewport rather than a whole desktop, reading the page itself (its accessibility tree, elements, forms, and tabs) and adding element references, form input, tab management, download reporting, and opt-in file upload on top of screenshot-and-click control. Both toolsets are available for Claude Fable 5, Claude Mythos 5, Claude Opus 5, Claude Sonnet 5, and Claude Opus 4.8 on the Claude API. The Files API is out of beta on the Claude API. Requests to the /v1/files endpoints, and Messages API requests that reference an uploaded file, no longer require the files-api-2025-04-14 beta header. Requests sent without the header use the current response format: file expiration (set expires_in_seconds when you upload a file; file objects report expires_at), and page and next_page pagination plus an ids[] filter when you list files. /v1/files requests that still send the beta header keep working and return the previous response format. Agent Skills and the Skills API (/v1/skills) are out of beta on the Claude API. Requests no longer require the skills-2025-10-02 beta header, including Messages API requests that load Skills through the container parameter. Requests that still send the header continue to work unchanged. See Using Agent Skills with the API. The Admin API user-management endpoints for Claude Enterprise (claude.ai) organizations (members, invites, groups, and custom roles) are out of beta. The anthropic-beta: ce-user-management-2026-07-13 header is no longer required on group and custom-role requests; requests that still send it are accepted unchanged. See User management. You can now restrict which sites a Claude Managed Agents agent's web_search and web_fetch tools can reach. Set allowed_domains or blocked_domains on the tool's entry in the agent_toolset_20260401 configs array; web_fetch also accepts max_content_tokens and web_search accepts user_location. Each configs entry is identified by its name and typed by an optional type, and requests that pass only name, enabled, and permission_policy continue to work; in the typed SDKs, configs entries become per-tool types. See Restrict web search and web fetch domains. Claude Managed Agents sessions that run in a self-hosted sandbox can now attach memory stores. The Python, TypeScript, and Go SDK workers download each attached store into the sandbox at its mount_path and sync the agent's changes back to the store. See Use memory stores. The session viewer in the Claude Console has been redesigned with a timeline minimap, a transcript grouped by model request, and an Inspector panel for session details and cost, raw events, per-tool statistics, mounted resources, and per-thread activity. See Console observability.

中文翻译待同步，当前显示官方原文。

站内阅读

A new dedicated student hub, study notebooks on mobile, one year of Google AI Pro in the US, or Google AI Plus in 140+ countries at no cost.

A new dedicated student hub, study notebooks on mobile, one year of Google AI Pro in the US, or Google AI Plus in 140+ countries at no cost. What: We’re introducing a dedicated student hub in the Gemini app to help students easily manage classes and create personalized study notebooks from their own course materials like syllabi. Study notebooks—first launched on desktop in June—are now available on mobile to help students get organized, prepare for class and stay on top of their material. Plus, eligible college students can get one year of Google AI Pro in the US, or Google AI Plus in 140+ countries at no cost. Why: By bringing study notebooks to mobile, creating a centralized student hub, and providing a student plan, we are helping students learn with a personalized study partner. Eligible students only. Offer Terms apply. Redeem until December 31, 2026. Automatically renews at the standard monthly price for the applicable Google AI Plan included in your Offer. Cancel anytime.

中文翻译待同步，当前显示官方原文。

站内阅读

ChatGPT Ads expands across Europe

ChatGPT Ads is expanding to 31 European markets. Learn how advertisers can reach people as they explore, compare options, and make decisions.

中文翻译待同步，当前显示官方原文。

站内阅读

Pacing model development in an era of cyber-critical capabilities

OpenAI is strengthening monitoring, alignment, and security for frontier AI models. See how new safeguards are guiding the pace of model development.

中文翻译待同步，当前显示官方原文。

站内阅读

Introducing ChatGPT for Teens: Built for learning, backed by protections

ChatGPT for Teens helps teens learn, think critically, and use AI with confidence, with stronger built-in protections, healthy-use features, and additional controls for parents.

中文翻译待同步，当前显示官方原文。

站内阅读

Asana completed a years-long code migration in 2 weeks with Codex

Asana used OpenAI Codex to replace an outdated testing system in two weeks, with about $12K in model and infrastructure costs.

中文翻译待同步，当前显示官方原文。

站内阅读

How NVIDIA scales expertise with ChatGPT Work

NVIDIA teams use ChatGPT Work to reduce manual tasks, connect fast-moving signals, and scale successful workflows globally.

中文翻译待同步，当前显示官方原文。

站内阅读

Workbench is now playground in the Claude Console.

Workbench is now playground in the Claude Console. Playground supports every Messages API parameter and includes templates that demonstrate API features such as code execution and web search. It shows the full SDK request and the API response for each run, to help you understand the API and build with it. For more, see the Claude Help Center or try it at platform.claude.com/playground.

中文翻译待同步，当前显示官方原文。

站内阅读

推出 Gemini 3.7 Flash

站内阅读

GPT-5.6 构建者指南

了解初创公司如何利用 GPT-5.6 通过更智能的模型选择和新的 Responses API 功能，构建更快、更具成本效益的 AI 代理。

站内阅读

预览超快模式：GPT-5.6 Sol 速度提升高达 14 倍

预览超快模式，这是 OpenAI API 的一个新服务层级，运行 GPT-5.6 Sol 的速度最高可提升 14 倍。由 Cerebras 提供支持，每秒可输出多达 750 个 token。

站内阅读

Gemini 3.7 Flash 正式发布（GA）：推出我们迄今为止最智能的编码和智能体工作主力模型：Gemini 3.7 Flash（gemini-3.7-flash）

Gemini 3.7 Flash 正式发布（GA）：推出我们迄今为止最智能的编码和智能体工作主力模型：Gemini 3.7 Flash（gemini-3.7-flash）：在软件工程、Web 开发和智能体工作流方面均有显著改进，并在 2026 年 12 月 31 日前提供优惠价格。要开始使用，请参阅 Gemini 3.7 Flash 模型页面和最新模型指南。

站内阅读

DeepSeek-V4-Pro 更新​

DeepSeek-V4-Pro 更新​ DeepSeek-V4-Pro 正式版已同步在 APP、网页端和 API 更新上线。API 调用方式不变，模型名设置为 deepseek-v4-pro，即可使用最新版本。 Agent 能力大幅提升 正式版 DeepSeek V4 Pro 极大增强了 Agent 能力，在生产环境中的性能表现提升尤为显著。 HLE (wo / w tools): 42.7/60.0 Terminal Bench 2.1: 87.9 NL2Repo: 61.5 Cybergym: 83.3 DeepSWE: 62.7 Toolathlon-Verified: 74.1 Agents' Last Exam: 25.7 AutomationBench (Public): 31.8 DSBench-FullStack: 71.1 DSBench-Hard: 67.2 原生支持 Responses API DeepSeek API 现已原生支持 OpenAI Responses API 格式，并针对性适配 Codex。用户可以参考官方文档，通过一键配置脚本完成 Codex 的配置。 更灵活的思考强度控制 V4-Pro 和 V4-Flash 思考模式现支持 low / high / max 三档思考强度，用户在实际使用中可以根据任务复杂度灵活选择：简单任务使用 low，日常 Agent 任务使用 high，更复杂的任务场景使用 max。设置方法请参考官方 API 文档：思考模式。 API 定价调整 随着 DeepSeek V4 全系列模型正式版上线，我们将对 API 价格进行更新调整。为了更加合理地调配资源，我们将采用峰谷定价，闲时价格为高峰时段价格的一半，鼓励用户根据实际使用情况调整任务时间。新价格将于北京时间 2026 年 8 月 17 日 0 时开始生效。 详细更新内容请参阅文档

站内阅读

从辅助到执行：企业如何让AI投入工作

OpenAI的研究揭示了企业如何采用代理式AI，使用ChatGPT和Codex，以及前沿企业如何在AI采用方面领先。

站内阅读

从工程到运维，RingCentral 如何构建 AI 原生化工作

了解 RingCentral 如何利用 ChatGPT Work 和 Codex 加速 AI 产品开发，并在工程和运维中集中运营智能。

站内阅读

Daybreak 模型现已在 AWS 上可用

OpenAI 和 AWS 正在通过 Amazon Bedrock 提供 Daybreak 网络安全能力，以支持企业安全工作流程。

站内阅读

在ChatGPT中测试广告

OpenAI开始在ChatGPT中测试广告，以支持免费访问，广告有明确标识、答案独立性、强大的隐私保护和用户控制。

站内阅读

合规API现可返回在用户机器上运行的Cowork和Claude Code会话记录，面向Claude Enterprise组织提供测试版。

合规API现可返回在用户机器上运行的Cowork和Claude Code会话记录，面向Claude Enterprise组织提供测试版。GET /v1/compliance/apps/sessions/local列出组织内的所有会话，GET /v1/compliance/apps/sessions/local/{session_id}检索单个会话的元数据，GET /v1/compliance/apps/sessions/local/{session_id}/messages返回其记录，均使用现有的合规访问密钥和read:compliance_user_data作用域。参见检索本地会话。

站内阅读

Model ML借助GPT-5.6 Sol更高效完成金融工作

Model ML使用GPT-5.6 Sol处理金融工作，从研究和分析到可编辑、可追溯的PowerPoint演示文稿和Excel工作簿。

站内阅读

随着网络防御窗口收窄，Daybreak 不断扩展

认识 GPT-5.6-Cyber，这是 OpenAI 专为网络安全设计的模型，可通过 Daybreak Red 用于授权漏洞研究、漏洞验证和安全测试。

站内阅读

将前沿网络模型交予更值得信赖的机构

经批准的Daybreak合作伙伴可以使用OpenAI的前沿网络模型，为客户提供经授权且受监管的网络安全服务。

站内阅读

Zapier如何利用ChatGPT Work转型核心营销流程

Zapier的企业营销团队使用ChatGPT Work来减少潜在客户漏斗中的流失数量，构建活动资产，并自动化报告。

站内阅读

ChatGPT Business 即将推出高级席位

ChatGPT Business 即将推出高级席位。在8月20日前注册，即可获得100美元的工作区积分，并为团队最繁重的工作解锁更高的使用额度。

站内阅读

维珍航空借助ChatGPT Work优化客户旅程

维珍航空正在利用ChatGPT Work加速研究、产品规划和决策，帮助团队连接客户旅程中的各种信号。

站内阅读

Claude Sonnet 5 的入门定价（每百万 Tok 2 美元 / 10 美元）现已成为标准价格：原定于 2026 年 9 月 1 日上调至每百万 Tok 3 美元 / 15 美元的计划将不会实施

Claude Sonnet 5 的入门定价（每百万 Tok 2 美元 / 10 美元）现已成为标准价格：原定于 2026 年 9 月 1 日上调至每百万 Tok 3 美元 / 15 美元的计划将不会实施。详见定价页面。

站内阅读

HSP GRUPPE如何为税务咨询构建AI能力

了解HSP GRUPPE如何利用ChatGPT Enterprise提升生产力、改善工作质量，并为税务咨询和客户服务创造更多容量。

站内阅读

您现在可以为Claude托管代理会话设置预算：会话支出的硬性上限，按公开列表价格计费。

您现在可以为Claude托管代理会话设置预算：会话支出的硬性上限，按公开列表价格计费。达到预算的会话会以budget_reached停止原因暂停，而不是启动新的模型请求；更改或移除预算可恢复会话。部署接受相同的预算，并将其应用于它们启动的每个会话。请参阅会话预算。您现在可以为Claude托管代理会话指定一个顾问：一个至少与代理自身能力相当的模型，会话的主线程可以在中途咨询该模型以获取战略指导。在代理的多代理名册中将其配置为{"type": "advisor"}条目，并指定要咨询的模型。请参阅为会话指定顾问。您现在可以控制Claude托管代理代理的模型推理运行位置。在创建代理时，在模型对象中设置inference_geo，或为单个会话覆盖它。请参阅数据驻留以了解可用的地理位置和定价。Claude托管代理会话现在可以从GitHub仓库加载技能。当会话挂载仓库时，其根目录.claude/skills目录中的任何技能都会在会话开始时自动发现，并在该会话中供代理使用。

站内阅读

在ChatGPT中改进GPT-5.6 Sol——并扩大免费用户对GPT-5.6 Luna的访问

ChatGPT推出了改进的GPT-5.6 Sol，具有更高的准确性和一致性，同时扩大了免费用户的访问权限，并提供了对GPT-5.6 Luna的无限制日常聊天。

站内阅读

从提问到行动：全球如何让ChatGPT投入工作

新的OpenAI Signals数据显示了全球人们如何使用ChatGPT，提供了关于采用率、使用趋势和不断演变的行为的国家级洞察。

站内阅读

技能与插件安全扫描（测试版）企业版计划现在可以开启技能与插件安全扫描，以自动检查第三方技能和插件是否存在恶意内容

技能与插件安全扫描（测试版）企业版计划现在可以开启技能与插件安全扫描，当有人上传或编辑第三方技能和插件时，自动检查其是否包含恶意内容。更多信息，请参阅技能与插件扫描入门。

站内阅读

推理钩子现已面向Claude Enterprise组织推出测试版。

推理钩子现已面向Claude Enterprise组织推出测试版。将Claude指向您组织的AI安全服务器，claude.ai、Cowork和Claude Code上的每个受治理提示都会在推理进行前被服务器拦截，以做出允许或拒绝的裁决。请求经过签名，故障处理可配置，每次拒绝都会记录在合规性活动源中。请参阅推理钩子。我们已退役Claude Opus 4.1模型（claude-opus-4-1-20250805）。对该模型的所有请求现在都将返回错误。我们建议升级到Claude Opus 5。研究人员可通过外部研究人员访问计划申请持续访问权限。

站内阅读

涉及OpenAI模型的第三方网络评估

OpenAI解释了近期第三方网络安全评估事件，并概述了加强AI模型测试与评估的新保障措施。

站内阅读

使用ChatGPT Work和Codex学习与教学的新方式

探索ChatGPT Work和Codex的新教育插件，帮助K-12教师、大学教育者和学生学习、教学、研究和构建。

站内阅读

我们如何在六个月内构建响应式语音AI的实时系统

GPT-Live实现了与AI的连续语音交互，采用无轮次语音模型和低延迟架构，以实现更快、更自然的对话。

站内阅读

Circles借助OpenAI技术推动电信个性化服务

Circles利用OpenAI API和Codex驱动AI原生的电信体验，使ARPU提升22%，客户流失率降低9%，并提高了开发效率。

站内阅读

合规API现可返回在claude.ai网页或移动端发起的Cowork会话记录，面向Claude Enterprise组织提供测试版。

合规API现可返回在claude.ai网页或移动端发起的Cowork会话记录，面向Claude Enterprise组织提供测试版。GET /v1/compliance/apps/sessions/remote列出会话，GET /v1/compliance/apps/sessions/remote/{session_id}/messages返回单个会话的记录，使用现有的合规访问密钥及read:compliance_user_data权限范围。参见检索远程会话。

站内阅读

Dreams（研究预览版）现已支持Claude Opus 5。

Dreams（研究预览版）现已支持Claude Opus 5。请参阅支持的模型。

站内阅读

Univé打造AI就绪的员工队伍

了解Univé如何通过结合领导力、负责任治理和员工主导的创新，利用ChatGPT Enterprise打造AI就绪的员工队伍，实现大规模工作转型。

站内阅读

Disrupting a Cambodia-based criminal scam operation

OpenAI disrupted a Cambodia-based scam operation using ChatGPT to support investment, romance, gambling, and impersonation schemes.

中文翻译待同步，当前显示官方原文。

站内阅读

DeepSeek-V4-Flash 更新​

DeepSeek-V4-Flash 更新​ DeepSeek-V4-Flash 正式版 API 上线公测，API 调用方式不变，模型名设置为 deepseek-v4-flash 即可使用最新版本。 Agent 能力大幅增强，基准测试远超 V4-Pro-Preview： Terminal Bench 2.1: 82.7 NL2Repo: 54.2 Cybergym: 76.7 DeepSWE: 54.4 Toolathlon verified: 70.3 Agent Last Exam: 25.2 Automation Bench (Public): 25.1 DSBench-FullStack: 68.7 DSBench-Hard: 59.6 注1：对于公开基准测试集中的 Code Agent 任务，正式版 DeepSeek-V4-Flash 使用 DeepSeek Harness 极简模式（即将发布）作为框架进行测试，并使用 max 档位，topp=0.95，temperature=1.0 注2：DSBench-FullStack 是内部使用的全栈开发测试集，DSBench-Hard 是内部使用的 Coding Agent 难题测试集 正式版 V4-Flash 原生支持 Responses API 格式并针对性适配 Codex，具体配置方法请参考文档 DeepSeek-V4-Flash-0731 的模型结构、尺寸和 DeepSeek-V4-Flash-Preview 保持一致，仅重新进行了后训练。 注意：本次仅升级了 DeepSeek-V4-Flash 的 API 接口，DeepSeek-V4-Pro API 及 APP/WEB 端模型未做更改。 DeepSeek-V4-Pro 正式版将会尽快发布。

站内阅读

Gemini Robotics ER 2：以视频理解、任务编排和多机器人协作为机器人赋能

Gemini Robotics ER 2 帮助机器人进行推理、协作并解决现实世界任务。它在视频理解、工具编排和多机器人协作方面实现了质的飞跃，适用于机器人应用。

站内阅读

推进GPT-5.6的价格性能前沿

探索Luna和Terra更低的GPT-5.6定价——以及OpenAI更高效的模型如何帮助企业大规模部署AI工作流。

站内阅读

avatarin 如何利用 GPT-Realtime 构建 24/7 零售代理

avatarin 使用 OpenAI 的 GPT-Realtime 为山田电机购物者提供 24/7 多语言支持。两周内，3万人使用了该代理，92% 的调查回复为正面。

站内阅读

Gemini Robotics ER 2 公开预览：发布两个用于机器人技术的新具身推理模型端点

Gemini Robotics ER 2 公开预览：发布了两个用于机器人技术的新具身推理模型端点：gemini-robotics-er-2-preview：高级空间推理、代理代码执行、多步骤工具编排、视频时刻查找、进度分类和多机器人协调。gemini-robotics-er-2-streaming-preview：针对使用 Live API 的实时文本流进行了优化，实现具有双向音频和视频输入的低延迟机器人代理。两个模型端点都接受文本、图像、视频和音频输入，并支持具有阻塞行为的函数调用以执行物理机器人动作。要开始使用，请参阅 Gemini Robotics ER 概述。对于实时流用例，请参阅带流式传输的机器人技术。弃用公告：gemini-robotics-er-1.6-preview 模型将于 2026 年 8 月 31 日关闭。

站内阅读

启用两个设置如何使我们在ARC-AGI-3基准测试中的得分翻了三倍

两个API设置如何通过保留推理和启用压缩来提升GPT-5.6在ARC-AGI-3上的性能，从而提高得分和效率。

站内阅读

利用ChatGPT加速学术研究者的科学发现

OpenAI向10万名学术研究者免费提供ChatGPT最先进的AI模型，以加速科学研究、合作与发现。

站内阅读

GPT-5.6如何融合前沿智能与前沿效率

GPT-5.6在模型、推理和智能体工作流方面提升了AI效率，帮助每美元提供更多有用的智能。

站内阅读

在 macOS 上使用 Gemini 通过语音在桌面任意位置创建和编辑

我们推出了一种仅用语音与 Gemini 协作的新方式，就在你所在之处。按住 Fn 键即可在活动窗口中口述，Gemini 会将你的语音转录为干净、精炼的文本，直接出现在光标处。你还可以在应用设置中启用 Gemini 推理，基于屏幕上下文执行更多操作——例如高亮文本进行重写或重新格式化、选择本地文件提取信息，或在任意位置生成图像。该新功能今日起以英语向所有 Gemini macOS 应用用户推出。在 gemini.google/mac 下载应用。为什么：说出你的想法让你以思维的速度工作——允许你漫谈、犹豫或自我修正，而 Gemini 无缝理解你的意图。通过获取屏幕上下文并将结果直接输出到任意应用的光标处，Gemini 与你并肩工作，让你保持流畅状态。

站内阅读

Gemini Robotics 2 为机器人带来全身智能

站内阅读

人工智能如何扩展人们的工作内容

OpenAI的新研究显示，人工智能正在扩展员工的工作内容，ChatGPT用户跨角色承担任务，重塑了工作边界。

站内阅读

Claude Opus 5 发布

我们发布了 Claude Opus 5，这是一款深思熟虑且主动的模型，其智能水平接近 Claude Fable 5 的前沿水平，但价格仅为后者的一半。

站内阅读

我们发布了Claude Opus 5（claude-opus-5），相比Claude Opus 4.8有显著改进。

我们发布了Claude Opus 5（claude-opus-5），相比Claude Opus 4.8有显著改进。Claude Opus 5支持100万token上下文窗口（默认和最大值均为100万）、12.8万最大输出token，并默认开启思考功能，定价为每百万token $5 / $25，与Claude Opus 4.8相同。该模型已在Claude API、Amazon Bedrock上的Claude、Google Cloud上的Claude以及Microsoft Foundry上的Claude中可用。新功能、行为变更和迁移指南请参阅《Claude Opus 5的新特性》，完整规格请参阅模型概述。在Claude Opus 5上，仅允许在effort为high或以下时禁用思考：使用effort xhigh或max时，thinking: {"type": "disabled"}会返回400错误，这是与Claude Opus 4.8的破坏性变更。详情见《Claude Opus 5的新特性》。Effort是引导Claude Opus 5的主要控制方式：模型支持完整的阶梯（low、medium、high、xhigh、max），其中max用于能力关键型工作。对话中途工具变更现已在Claude Fable 5、Claude Mythos 5、Claude Opus 4.8和Claude Opus 5上进入测试阶段：可在对话轮次之间添加或移除工具，同时保留提示缓存。请在请求中包含mid-conversation-tool-changes-2026-07-01测试版标头。fallbacks参数现在支持“default”模式，该模式根据拒绝类别应用Anthropic推荐的备用模型。服务器端备用功能处于测试阶段，“default”模式需要server-side-fallback-2026-07-01测试版标头。详见《拒绝与备用》。我们已移除Claude Opus 4.7的快速模式。对claude-opus-4-7使用speed: "fast"的请求现在会返回错误；与Claude Opus 4.6不同，它们不会回退到标准速度。Claude Opus 4.7本身仍以标准速度可用。要继续使用快速模式，请迁移到Claude Opus 5或Claude Opus 4.8。更多信息请阅读《快速模式》。

站内阅读

在ChatGPT中推出健康功能

ChatGPT中的健康功能现在允许符合条件的美国用户安全地连接医疗记录和Apple Health，以获得更个性化的见解并更好地了解自己的健康状况。

站内阅读

与埃芬汉县社区共建AI基础设施

OpenAI宣布在佐治亚州埃芬汉县启动“卡米莉亚项目”，承诺负责任能源、社区投资、就业及Codex访问权限。

站内阅读

OpenAI Presence 介绍

介绍 OpenAI Presence，一个经过验证的企业级 AI 代理平台，帮助组织部署可信的语音和聊天代理，用于客户和内部工作流程。

站内阅读

NTT DATA集团利用Codex将事件分析时间缩短至30分钟

NTT DATA集团使用ChatGPT Enterprise和Codex帮助9000名员工实现工作自动化，将事件分析时间缩短至30分钟，并扩大安全AI的应用规模。

站内阅读

你现在可以在Claude托管代理的模型配置中设置努力级别。

你现在可以在Claude托管代理的模型配置中设置努力级别。创建代理时，在模型对象中传递effort参数。查看努力级别了解每个级别的作用。Claude托管代理的Webhooks现在覆盖环境和记忆存储生命周期：四个environment.*事件类型和三个memory_store.*事件类型。你可以无需轮询即可响应环境和记忆存储生命周期变化。请参阅订阅Webhooks中的环境事件和记忆存储事件选项卡。创建Claude托管代理会话时，你现在可以用初始事件进行种子化。在POST /v1/sessions中传递initial_events，最多包含50个user.message和user.define_outcome事件。非空列表会在同一调用中启动代理循环，因此你无需单独的发送事件请求来开始工作。更新Claude托管代理时，version字段现在是可选的。提供它以进行乐观并发控制（不匹配返回409错误），或省略它以无条件应用更新。请参阅更新语义。Claude托管代理会话线程事件流现在支持事件增量。GET /v1/sessions/{session_id}/threads/{thread_id}/stream接受与会话级流相同的event_deltas[]查询参数，因此你可以在模型生成时预览子代理的文本。连接仅预览它正在读取的线程。请参阅预览会话线程事件。

站内阅读

为小企业推出ChatGPT计划

OpenAI推出小企业ChatGPT计划，帮助创业者培养AI技能、自动化工作，并借助ChatGPT Work实现增长。

站内阅读

推出 Gemini 3.6 Flash、3.5 Flash-Lite 和 3.5 Flash Cyber

我们正在推出新的 Gemini 模型，包括 Gemini 3.6 Flash、3.5 Flash-Lite 和 3.5 Flash Cyber。

站内阅读

OpenAI与Hugging Face合作应对模型评估期间的安全事件

OpenAI和Hugging Face分享了AI模型评估期间一起安全事件的初步发现，凸显了先进的网络能力及对防御者的启示。

站内阅读

3.6 Flash：为您日常生活升级的模型

我们推出了 Gemini 3.6 Flash，这是 Flash 模型的升级版本，旨在帮助您处理日常任务。尝试将其用于快速任务，如分析多个文档，或多步骤项目，如使用 Canvas 工具构建工作原型。我们很高兴将此模型带给全球所有 Gemini 应用用户。要访问，请在模型下拉菜单中选择“3.6 Flash”。我们希望让 Gemini 成为对每个人都最有帮助的 AI 助手。借助 Gemini 3.6 Flash，您现在可以更自信地处理日常任务，无需在速度和质量之间做出选择。

站内阅读

Gemini 3.6 Flash 和 Gemini 3.5 Flash-Lite 正式发布（GA）：发布了我们最新的 3.x Flash 模型的稳定、生产就绪版本：Gemini 3.6 Flash (gemini-3.6-flash)：

Gemini 3.6 Flash 和 Gemini 3.5 Flash-Lite 正式发布（GA）：发布了我们最新的 3.x Flash 模型的稳定、生产就绪版本：Gemini 3.6 Flash (gemini-3.6-flash)：改进了 token 效率以及代码/智能体规划能力，价格低于 3.5 Flash，解决了开发者关于输出冗长的反馈。Gemini 3.5 Flash-Lite (gemini-3.5-flash-lite)：提供低延迟、高性价比的子智能体选项，专为高容量自动化设计。更多信息请参阅最新的 Gemini 模型指南。已弃用的参数：采样参数 temperature、top_p 和 top_k 现已弃用。详情请参阅最新的 Gemini 模型。

站内阅读

长周期模型时代的安全与对齐

OpenAI分享了部署长时间运行AI模型的经验，强调了新的安全风险、观察到的失败案例以及通过迭代部署改进的安全措施。

站内阅读

介绍 Gemini 3.5 Flash Cyber

谷歌推出 Gemini 3.5 Flash Cyber，一款轻量级网络安全模型，用于发现和修补漏洞。

站内阅读

Claude 控制台中的旧版 Workbench (platform.claude.com/workbench) 即将停用，访问权限将于 2026 年 8 月 17 日结束。

Claude 控制台中的旧版 Workbench (platform.claude.com/workbench) 即将停用，访问权限将于 2026 年 8 月 17 日结束。更新的 Workbench 不支持保存的提示词、变量和评估。您可以从横幅和“组织设置”下导出想要保留的任何数据。更多信息，请参阅 Claude 帮助中心的“如何使用 Workbench？”。用于生成、改进和模板化提示词的实验性提示工具 API（/v1/experimental/generate_prompt、/v1/experimental/improve_prompt 和 /v1/experimental/templatize_prompt）将于 2026 年 8 月 17 日与 Workbench 一同退役。移除后，对这些端点的请求将返回错误。

站内阅读

为什么青少年应该获得安全的AI

了解OpenAI如何通过适龄保护、学习工具、家长控制和专家合作，让ChatGPT对青少年更安全。

站内阅读

Codex 如何成为 OpenAI 创意团队的协作者

OpenAI 创意团队如何使用 Codex 构建定制创意工具、加速构思，并借助上下文感知 AI 更快地制作原型。

站内阅读

GPT-Red：解锁鲁棒性的自我改进

探索GPT-Red，OpenAI的自动化红队系统，通过自我对弈提升AI安全性、对齐性和提示注入鲁棒性。

站内阅读

对话中系统消息现可用于Claude Fable 5、Claude Mythos 5和Claude Opus 4.8，在Claude API、Amazon Bedrock上的Claude以及Google Cloud上。

对话中系统消息现可用于Claude Fable 5、Claude Mythos 5和Claude Opus 4.8，在Claude API、Amazon Bedrock上的Claude以及Google Cloud上。无需beta标头。这修正了先前的可用性说明。

站内阅读

Sales workflows with ChatGPT Work

Learn practical ChatGPT Work workflows for pipeline briefs, meeting prep, account plans, forecast reviews, and deal diagnosis.

中文翻译待同步，当前显示官方原文。

站内阅读

Data science workflows with ChatGPT Work

Learn practical ChatGPT Work workflows for root-cause briefs, KPI memos, scoped analyses, and dashboard specifications.

中文翻译待同步，当前显示官方原文。

站内阅读

您的Claude组织的HIPAA配置现已自助服务

您现在可以自行管理Claude组织的HIPAA就绪状态。这适用于Claude Enterprise和Claude平台（API）。在每个产品中，符合条件的管理员可以审查业务伙伴协议（BAA）、下载实施指南，并在单一流程中启用HIPAA配置。有关更多信息，请参阅HIPAA就绪企业计划和Claude API的HIPAA就绪状态。

站内阅读

您现在可以通过Admin API管理Claude Enterprise（claude.ai）组织中的人员，该API对所有Claude Enterprise组织处于测试阶段：列出成员并查找他们

您现在可以通过Admin API管理Claude Enterprise（claude.ai）组织中的人员，该API对所有Claude Enterprise组织处于测试阶段：列出成员并通过电子邮件地址查找、更改成员角色、移除成员、发送和撤销邀请、管理群组及其成员资格，以及读取自定义角色。群组和自定义角色请求需要anthropic-beta: ce-user-management-2026-07-13 beta标头；成员和邀请请求不需要beta标头。具有read:org_audit范围的Admin API密钥也可以调用所有用户管理的GET端点。请参阅用户管理。

站内阅读

用ATL Saathi赋能印度下一代创新者

谷歌和AIM推出了ATL Saathi，这是一款由Gemini驱动的AI工具，旨在赋能印度机器人实验室的教育工作者。

站内阅读

ChatGPT 入门指南

了解如何使用 ChatGPT，开始您的第一次对话，并发现利用 AI 进行写作、头脑风暴和解决问题的简单方法。

站内阅读

Claude记忆更新：现在作为一组独立、分类的条目工作

Claude上的记忆更新现在作为一组独立、分类的条目工作，Claude在对话中读取并更新这些条目，取代了之前的每日记忆摘要。请参阅使用Claude的聊天搜索和记忆来构建之前的上下文。

站内阅读

Dreams（研究预览版）现已支持 Claude Fable 5 和 Claude Sonnet 5。

Dreams（研究预览版）现已支持 Claude Fable 5 和 Claude Sonnet 5。请参阅支持的模型。我们扩展了 cmek_preserve 事件的访问透明度文档，增加了过滤示例、事件负载示例以及两个保留原因代码（policy_violation_investigation、csae_report）。文档现在还澄清，无论保留是由人工审核员还是自动化安全管道发起，都会写入保留事件。请参阅 CMEK 内容保留。

站内阅读

GPT-5.6 现已成为 Microsoft 365 Copilot 的首选模型

了解 GPT-5.6 如何以更强的 AI 能力驱动 Microsoft 365 Copilot，在 Word、Excel、PowerPoint、Chat 和 Cowork 中实现更快速、更高质量的工作。

站内阅读

GPT-5.5 生物漏洞赏金计划

关于 OpenAI 生物漏洞赏金计划的详细信息

站内阅读

GPT-5.6：随你雄心扩展的前沿智能

每个令牌带来更多智能，每美元带来更强性能，按需提供更多能力，应对你最艰巨的工作。

站内阅读

ChatGPT 现已成为您最具雄心工作的合作伙伴

ChatGPT Work 是一个可以跨应用和文件采取行动的智能体，能在需要时持续处理项目数小时，并将目标转化为完成的工作。

站内阅读

与Claude一起反思的新方式

你的月度回顾是设置>反思中的一项新功能，显示你花费时间的主题、最活跃的日期和高峰时段，以及关于你如何与Claude合作的观察。该功能在网页版和Claude Desktop上的免费、专业和最大计划中处于测试阶段，并且需要开启记忆功能。此外，设置>时间和专注让你设置可选的休息提醒和安静时段。更多信息，请参阅查看你的月度回顾和设置休息提醒和安静时段。

站内阅读

在编码评估中区分信号与噪声

OpenAI的一项新分析揭示了热门编码基准SWE-Bench Pro存在的问题，引发了对评估AI模型可靠性和准确性的担忧。

站内阅读

DeepSeek-V4​

DeepSeek-V4​ DeepSeek API 已支持 V4-Pro 与 V4-Flash，支持 OpenAI ChatCompletions 接口与 Anthropic 接口。访问新模型时，base_url 不变, model 参数需要改为 deepseek-v4-pro 或 deepseek-v4-flash。 旧有的 API 接口的两个模型名 deepseek-chat 与 deepseek-reasoner 将于三个月后（2026-07-24）停止使用。当前阶段内，这两个模型名分别指向 deepseek-v4-flash 的非思考模式与思考模式。 详细更新内容请参阅文档

站内阅读

DeepSeek-V3.2​ / DeepSeek-V3.2-Speciale​

DeepSeek-V3.2​ deepseek-chat 和 deepseek-reasoner 都已升级为 DeepSeek-V3.2. deepseek-chat 对应 DeepSeek-V3.2 的非思考模式 deepseek-reasoner 对应 DeepSeek-V3.2 的思考模式 DeepSeek-V3.2-Speciale​ 我们非正式部署了 DeepSeek-V3.2-Speciale 的 API 服务，API 用户可以通过设置 base_url="https://api.deepseek.com/v3.2_speciale_expires_on_20251215" 访问该模型。该模型 API 价格不变，只支持思考模式下的对话功能，不支持工具调用等功能，最大输出长度默认为 128K，支持时间截止至北京时间 2025-12-15 23:59。 详细更新内容请参阅文档

站内阅读