studioglobal
答案公共網址2026年4月20日32 來源

搜尋及事實查核:Claude Opus 4.7 同 Opus 4.6 真正差幾多?值唔值得升級?

如果你主要用 Opus 做寫 code、跑 agent workflow、睇 screenshot/文件圖像,我會傾向話值得升級;如果你主要係普通聊天、寫作、摘要,官方證據未足以證明有「大幅」普遍提升。[12][15] 先講最重要:Opus 4.7 同 Opus 4.6 嘅 API 標價一樣,兩者都係 $5 / MTok input、$25 / MTok output。[13] 核心規格其實好接近:4.7 保留咗同 4.6 一樣嘅...

如果你主要用 Opus 做寫 code、跑 agent workflow、睇 screenshot/文件圖像,我會傾向話值得升級;如果你主要係普通聊天、寫作、摘要,官方證據未足以證明有「大幅」普遍提升。[12][15] 先講最重要:Opus 4.7 同 Opus 4.6 嘅 API 標價一樣,兩者都係 $5 / MTok input、$25 / MTok output。[13] 核心規格其實好接近:4.7 保留咗同 4.6 一樣嘅 1M context window、128k max output,同埋同一套工具與平台功能。[10] 真正差異主要喺能力同工作方式,唔係價錢或 context size。Anthropic 官方直接話

重點

  • 如果你主要用 Opus 做寫 code、跑 agent workflow、睇 screenshot/文件圖像,我會傾向話值得升級;如果你主要係普通聊天、寫作、摘要,官方證據未足以證明有「大幅」普遍提升。[12][15]
  • 先講最重要:Opus 4.7 同 Opus 4.6 嘅 API 標價一樣,兩者都係 $5 / MTok input、$25 / MTok output。[13]

研究答案

如果你主要用 Opus 做寫 code、跑 agent workflow、睇 screenshot/文件圖像,我會傾向話值得升級;如果你主要係普通聊天、寫作、摘要,官方證據未足以證明有「大幅」普遍提升。[12][15]

  • 先講最重要:Opus 4.7 同 Opus 4.6 嘅 API 標價一樣,兩者都係 $5 / MTok input、$25 / MTok output。[13]
  • 核心規格其實好接近:4.7 保留咗同 4.6 一樣嘅 1M context window、128k max output,同埋同一套工具與平台功能。[10]
  • 真正差異主要喺能力同工作方式,唔係價錢或 context size。Anthropic 官方直接話 4.7 係佢哋最強嘅一般可用模型,並且喺 agentic coding 上對 4.6 有「step-change improvement」。[12]
  • 4.7 新增幾個實際有感嘅位:高解像圖片支援、xhigh effort level,同埋 task budgets(beta)。高解像圖片上限由之前 1568px / 1.15MP 提升到 2576px / 3.75MP,對 screenshot、文件、computer-use 類工作特別相關。[15]
  • 升級唔係完全冇代價。4.7 用咗新 tokenizer,官方話同一段固定文字有機會多用最多約 35% tokens,所以雖然「每百萬 token 單價」冇變,實際帳單未必真係一樣。[13][15]
  • API 遷移亦有 breaking changes:4.7 之後唔再支援舊式 extended thinking(
    i.j4i.i2
    thinking: {type: "enabled", budget_tokens: ...}
    ),要改用 adaptive thinking;另外 temperature、top_p、top_k 設成非預設值會報錯。[10]
  • 所以如果你而家嘅 4.6 workflow 係靠舊 thinking 參數、或者你好依賴手動 sampling 調校,升級前要先改 integration。[10]
  • 官方材料所展示嘅進步,重點集中喺長流程 agent work、複雜 coding、instruction following 同 vision;我未見到一個足以代表所有一般聊天/寫作/知識問答場景嘅中立公開總分,所以要話「全部場景都明顯勁好多」,證據仍然有限。[12][15]

我嘅實際建議:

  • 值得升級:

    • 你用佢做 coding agent、除 bug、重構、大型多步任務
    • 你會餵 screenshot、UI、PDF、圖像文件
    • 你想用最新能力,但又唔想多付每-token 單價
  • 可以暫緩:

    • 你主要做普通對話、文案、摘要
    • 你現有 4.6 prompt 已經好穩,升級成本高
    • 你對 token 成本好敏感,或者 workflow 依賴舊 thinking / sampling 設定

如果你想,我可以下一步幫你整一個「Opus 4.6 vs 4.7」30 分鐘 A/B 測試清單,專門用你自己嘅 prompt 去驗證升級值唔值。

支持視覺效果

The image features the text "Claude" in bold black font, accompanied by an orange starburst-like logo, with the words "BY ANTHROPIC" below in smaller gray letters, all set against
The image features the text "Claude" in bold black font, accompanied by an orange starburst-like logo, with the words "BY ANTHROPIC" below iThe image features the text "Claude" in bold black font, accompanied by an orange starburst-like logo, with the words "BY ANTHROPIC" below in smaller gray letters, all set against a light beige background.
The image features the logo and name of "Claude Opus 4.6," with a stylized flower-like symbol in a reddish color next to the text.
The image features the logo and name of "Claude Opus 4.6," with a stylized flower-like symbol in a reddish color next to the textThe image features the logo and name of "Claude Opus 4.6," with a stylized flower-like symbol in a reddish color next to the text.
Active Inference AI: Here's Why It's The Future of Enterprise Operations and Industry Innovation
Active Inference AI: Here's Why It's The Future of Enterprise Operations and Industry Innovation
Exploring the Spatial Web Protocol at the 4th Annual Applied Active Inference Symposium
Exploring the Spatial Web Protocol at the 4th Annual Applied Active Inference Symposium
featured image - Claude Opus 4.7 Is Here and It Changes the Coding Model Race
featured image - Claude Opus 4.7 Is Here and It Changes the Coding Model Race
The Trick That Agentic Frameworks Pulled On Us
The Trick That Agentic Frameworks Pulled On Us
Alex Cloudstar HackerNoon profile picture
Alex Cloudstar HackerNoon profile picture
Building the "Spotify Wrapped" of Product Hunt: A 48-Hour Technical Sprint to Map the Maker Economy
Building the "Spotify Wrapped" of Product Hunt: A 48-Hour Technical Sprint to Map the Maker Economy
The foundation model era of computer vision.
The foundation model era of computer visionThe foundation model era of computer vision.
Free data annotation guide book cover
Free data annotation guide book cover
A7coZ0uhYMgPNXXtkg7aiDfCsxT2 yd83ifc
A7coZ0uhYMgPNXXtkg7aiDfCsxT2 yd83ifc
top egocentric datasets
top egocentric datasets
Gemini Robotics-ER 1.6
Gemini Robotics-ER 1.6
Original Reporting
Original Reporting
Opus 4.7
Opus 4.7
avatar
avatar
Akash 6
Akash 6
caveman newsletter
caveman newsletter

研究對話

研究問題

搜尋及事實查核:Claude Opus 4.7 同 Opus 4.6 真正差幾多?值唔值得升級?

Studio Global AI32 來源

引用的答案

如果你主要用 Opus 做寫 code、跑 agent workflow、睇 screenshot/文件圖像,我會傾向話值得升級;如果你主要係普通聊天、寫作、摘要,官方證據未足以證明有「大幅」普遍提升。[12][15]

  • 先講最重要:Opus 4.7 同 Opus 4.6 嘅 API 標價一樣,兩者都係 $5 / MTok input、$25 / MTok output。[13]
  • 核心規格其實好接近:4.7 保留咗同 4.6 一樣嘅 1M context window、128k max output,同埋同一套工具與平台功能。[10]
  • 真正差異主要喺能力同工作方式,唔係價錢或 context size。Anthropic 官方直接話 4.7 係佢哋最強嘅一般可用模型,並且喺 agentic coding 上對 4.6 有「step-change improvement」。[12]
  • 4.7 新增幾個實際有感嘅位:高解像圖片支援、xhigh effort level,同埋 task budgets(beta)。高解像圖片上限由之前 1568px / 1.15MP 提升到 2576px / 3.75MP,對 screenshot、文件、computer-use 類工作特別相關。[15]
  • 升級唔係完全冇代價。4.7 用咗新 tokenizer,官方話同一段固定文字有機會多用最多約 35% tokens,所以雖然「每百萬 token 單價」冇變,實際帳單未必真係一樣。[13][15]
  • API 遷移亦有 breaking changes:4.7 之後唔再支援舊式 extended thinking(
    i.j4i.i2
    thinking: {type: "enabled", budget_tokens: ...}
    ),要改用 adaptive thinking;另外 temperature、top_p、top_k 設成非預設值會報錯。[10]
  • 所以如果你而家嘅 4.6 workflow 係靠舊 thinking 參數、或者你好依賴手動 sampling 調校,升級前要先改 integration。[10]
  • 官方材料所展示嘅進步,重點集中喺長流程 agent work、複雜 coding、instruction following 同 vision;我未見到一個足以代表所有一般聊天/寫作/知識問答場景嘅中立公開總分,所以要話「全部場景都明顯勁好多」,證據仍然有限。[12][15]

我嘅實際建議:

  • 值得升級:

    • 你用佢做 coding agent、除 bug、重構、大型多步任務
    • 你會餵 screenshot、UI、PDF、圖像文件
    • 你想用最新能力,但又唔想多付每-token 單價
  • 可以暫緩:

    • 你主要做普通對話、文案、摘要
    • 你現有 4.6 prompt 已經好穩,升級成本高
    • 你對 token 成本好敏感,或者 workflow 依賴舊 thinking / sampling 設定

如果你想,我可以下一步幫你整一個「Opus 4.6 vs 4.7」30 分鐘 A/B 測試清單,專門用你自己嘅 prompt 去驗證升級值唔值。

人們還問

「搜尋及事實查核:Claude Opus 4.7 同 Opus 4.6 真正差幾多?值唔值得升級?」的簡短答案是什麼?

如果你主要用 Opus 做寫 code、跑 agent workflow、睇 screenshot/文件圖像,我會傾向話值得升級;如果你主要係普通聊天、寫作、摘要,官方證據未足以證明有「大幅」普遍提升。[12][15]

首先要驗證的關鍵點是什麼?

如果你主要用 Opus 做寫 code、跑 agent workflow、睇 screenshot/文件圖像,我會傾向話值得升級;如果你主要係普通聊天、寫作、摘要,官方證據未足以證明有「大幅」普遍提升。[12][15] 先講最重要:Opus 4.7 同 Opus 4.6 嘅 API 標價一樣,兩者都係 $5 / MTok input、$25 / MTok output。[13]

接下來我應該探索哪個相關主題?

繼續“搜尋及事實查核:Claude Opus 4.7 同 ChatGPT / Gemini 比,邊個更啱我用?”以獲得另一個角度和額外的引用。

開啟相關頁面

我應該將其與什麼進行比較?

對照「搜尋及事實查核:香港會唔會有自己嘅大模型 / 本地 AI 生態?」交叉檢查此答案。

開啟相關頁面

繼續你的研究

來源

  • [1] What's new in Claude Opus 4.7platform.claude.com

    Claude Opus 4.7 introduces task budgets. This new tokenizer may use roughly 1x to 1.35x as many tokens when processing text compared to previous models (up to ~35% more, varying by content), and

    i.j4i.i2
    /v1/messages/count_tokens
    will return a different number of tokens for Claude Opus 4.7 than it did for Claude Opus 4.6. See [High-resolution image support](https://platform.claude.com/docs/en/about-claude/models/whats-new-claude-4-7#high-resolution-image-suppo…

  • [2] Claude (language model) - Wikipediaen.wikipedia.org
  • [3] [AINews] Anthropic Claude Opus 4.7 - literally one step better than ...latent.space

    Anthropic officially launched Claude Opus 4.7 as its newest top-tier Opus model, positioning it as better at long-running work, coding, instruction following, self-verification, computer use, and knowledge work than Opus 4.6, while keeping list pricing unchanged at $5 / $25 per million input/output tokens according to user summaries and launch discussion [@claudeai, @kimmonismus]. * Anthropic also launched or highlighted task budgets in public beta,

    i.j4i.i2
    /ultrareview
    in Claude Cod…

  • [4] Anthropic just released Opus 4.7 — the 'civilian' version of the AI ...tomsguide.com

    Unlock and manage exclusive Tom’s Guide member rewards. # Anthropic just released a 'Civilian' version of its 'Mythos' AI that's too dangerous for the public. Here at Tom’s Guide our expert editors are committed to bringing you the best news, reviews and guides to help you stay informed and ahead of the curve! Tom's AI Guide. Today, Anthropic officially released Claude Opus 4.7, the most powerful AI model available to the general public. To truly get why the release of Opus 4.7 is such a milestone, you first have to understand the implications of Anthropic's Claude Mythos Preview. Howeve…

  • [5] Claude Opus 4.7 - Anthropicanthropic.com
  • [6] Claude Opus 4.7 By Anthropic: Features, Updates & What You ...acecloud.ai

    Anthropic describes Claude Opus 4.7 as the most capable model for complex reasoning and agentic coding. Anthropic launched Claude Opus 4.7 on April 16, 2026, calling it its most capable generally available model, with major gains in advanced software engineering, instruction following, long-running tasks, and high-resolution vision. **Anthropicrecommendsit for long-horizon agentic work, knowledge work, vision tasks, and memory-heavy tasks.**That makes it especially relevant for developers, product teams, analysts, legal and finance workflows, AI startups, research teams, and enterpris…

  • [7] Claude Opus 4.7 vs Opus 4.6: features, performance, context ...datastudios.org

    Claude Opus 4.6 already arrived with a serious premium position inside Anthropic’s lineup, and it was introduced as a model designed for harder coding, more sustained agentic tasks, and longer multi-step reasoning chains that needed more persistence than lighter models could offer. Claude Opus 4.7 inherits that role, while pushing it further into a more explicit flagship profile that centers not only on difficult code and long tasks, but also on high-resolution vision, stronger knowledge work, more literal prompt handling, and a more tightly controlled API surface. Claude Opus 4.6 already pla…

  • [8] Claude Opus 4.7 vs Opus 4.6: What Actually Changed?labellerr.com

    Claude Opus 4.7 delivers major upgrades in coding, vision, and instruction precision. Anthropic released Claude Opus 4.7 on April 16, 2026. ## What Is Claude Opus 4.7 ? The model ID is

    i.j4i.i2
    claude-opus-4-7
    . ## New Features in Claude Opus 4.7. The model ID is
    i.j4i.i2
    claude-opus-4-7
    . Box's evaluations found that Opus 4.7 had a 56% reduction in model calls and a 50% reduction in tool calls compared to Opus 4.6 on certain workflows, meaning the model often gets tasks done in fewer steps, which can offset the higher per-token count. Claude is the model that developers and enterprises reach for when they n…

  • [9] Introducing Claude Opus 4.7 - Anthropicanthropic.com

    Skip to main contentSkip to footer. Developers can use

    i.j4i.i2
    claude-opus-4-7
    via the Claude API. Image 3: logo. Image 4: logo. ![Image 5: logo](https://www-cdn.anthropic.com/images/4zrzovbb/websi…

  • [10] Anthropic releases a new Opus model amid Mythos Preview buzztheverge.com

    Anthropic releases a new Opus model amid Mythos Preview buzz. Opus 4.7 comes on the heels of Mythos Preview, the buzzy cybersecurity-focused model Anthropic announced earlier this month, which the company has said is its most powerful model overall. In Opus 4.7’s system card, Anthropic wrote that Opus 4.7 doesn’t even advance the company’s “capability frontier,” since Claude Mythos Preview received higher results “on every relevant evaluation.”. “We stated that we would keep Claude Mythos Preview’s release limited and test new cyber safeguards on less capable models first,” Anthropic wrote…

  • [11] Claude Opus 4.7 Is Here and It Changes the Coding Model Racehackernoon.com

    Claude Opus 4.7 Is Here and It Changes the Coding Model Race | HackerNoon. Image 1Image 2: Hackernoon logo Hackernoon. [Image 3Image 4: Catchpoint Modern API Monitoring for your apps](https://www.catchpoint.com/application-experience/api-monitoring/?utm_campaign=Hackernoon-TOFU-billboard&utm_source=…

  • [12] Introducing Claude Opus 4.7, our most capable ...reddit.com

    Skip to main contentIntroducing Claude Opus 4.7, our most capable Opus model yet. Open menu Open navigationGo to Reddit Home. Get App Get the Reddit app Log InLog in to Reddit. Image 1 Go to ClaudeAI

  • [13] Claude Platform - Claude API Docsplatform.claude.com
  • [14] Claude Platform - Claude API Docsplatform.claude.com

    Cookie settings. We use cookies to deliver and improve our services, analyze site usage, and if you agree, to customize or personalize your experience and market our services to you. You can read our Cookie Policy here. ### Solutions. ### Partners. ### Learn. ### Company. ### Learn. ### Help and security. ### Terms and policies. # Claude Platform. Versionshinweise zu Claude Apps finden Sie in den Versionshinweisen für Claude Apps im Claude Help Center. Aktualisierungen zu Claude Code finden Sie im vollständigen CHANGELOG.md im
    i.j4i.i2
    claude-code
    -Repository.
    i.j4i.i2
    managed-agents-2026-04-01
    . `output…

  • [15] Migration guide - Claude API Docsplatform.claude.com

    It supports the same set of features as Claude Opus 4.6, including the 1M token context window at standard API pricing with no long-context premium, 128k max output tokens, adaptive thinking, prompt caching, batch processing, the Files API, PDF support, vision, and the full set of server-side and client-side tools (bash, code execution, computer use, text editor, web search, web fetch, MCP connector, memory). Extended thinking removed:

    i.j4i.i2
    thinking: {type: "enabled", budget_tokens: N}
    is no longer supported on Claude Opus 4.7 or later models and returns a 400 error. **Migrate to adaptive th…

  • [16] Pricing - Claude API Docsplatform.claude.com

    The "Base Input Tokens" column shows standard input pricing, "Cache Writes" and "Cache Hits" are specific to prompt caching, and "Output Tokens" shows output pricing. * Prompt caching multipliers apply on top of fast mode pricing. Fast mode is not available with the Batch API. Claude Mythos Preview, Opus 4.7, Opus 4.6, and Sonnet 4.6 in…

  • [17] Claude on Amazon Bedrock - Claude API Docsplatform.claude.com

    A research preview of a new AWS-managed offering, with the Messages API at

    i.j4i.i2
    /anthropic/v1/messages
    and SSE streaming, is documented at Claude in Amazon Bedrock (research preview). Calling Claude through Bedrock slightly differs from how you would call Claude when using Anthropic's client SDKs. This guide walks you through completing an API call to Claude on Bedrock using one of Anthropic's client SDKs. Note that this guide assumes you have already signed up for an AWS account and configured programmatic access. Go to the AWS Console > Bedrock > Model Access and request access to Anthropic mo…

  • [18] Claude on Amazon Bedrock - Claude API Docsplatform.claude.com

    A research preview of a new AWS-managed offering, with the Messages API at

    i.j4i.i2
    /anthropic/v1/messages
    and SSE streaming, is documented at Claude in Amazon Bedrock (research preview). Calling Claude through Bedrock slightly differs from how you would call Claude when using Anthropic's client SDKs. This guide walks you through completing an API call to Claude on Bedrock using one of Anthropic's client SDKs. Note that this guide assumes you have already signed up for an AWS account and configured programmatic access. Go to the AWS Console > Bedrock > Model Access and request access to Anthropic mo…

  • [19] Claude Platform - Claude API Docsplatform.claude.com

    Cookie settings. We use cookies to deliver and improve our services, analyze site usage, and if you agree, to customize or personalize your experience and market our services to you. You can read our Cookie Policy here. ### Terms and policies. Per le note sulla versione di Claude Apps, consulta le Note sulla versione di Claude Apps nel Centro assistenza Claude. Per gli aggiornamenti a Claude Code, consulta il CHANGELOG.md completo nel repository
    i.j4i.i2
    claude-code
    .
    i.j4i.i2
    advisor-tool-2026-03-01
    .
    i.j4i.i2
    output-300k-2026-03-24
    .
    i.j4i.i2
    claude-3-7-sonnet-20250219
    .
    i.j4i.i2
    claude-3-5-haiku-20241022
    . ### 24 novembre 20…

  • [20] Context windows - Claude API Docsplatform.claude.com

    The context window with extended thinking and tool use. When using extended thinking, all input and output tokens, including the tokens used for thinking, count toward the context window limit, with a few nuances in multi-turn situations. However, previous thinking blocks are automatically stripped from the context window calculation by the Claude API and are not part of the conversation history that the model "sees" for subsequent turns, preserving token capacity for actual conversation content. Context window diagram with extended thinking and tool use. Context window diagram with extend…

  • [21] Features overview - Claude API Docsplatform.claude.com

    | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedr…

  • [22] Features overview - Claude API Docsplatform.claude.com

    | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedrock (GA) Google Cloud's Vertex AI (GA) Microsoft Foundry (Beta) |. | ZDR eligible | Claude API (GA) Amazon Bedr…

  • [23] Models overview - Claude API Docsplatform.claude.com

    . * Build. * Admin. * Models & pricing. * Client SDKs. * API Reference. Pricing. Console. Once you've picked a model, [learn how to make your first API call](https://p…

  • [24] Pricing - Claude API Docsplatform.claude.com

    Prompt caching uses the following pricing multipliers relative to base input token rates:. For Claude Opus 4.7, Claude Opus 4.6, and newer models, specifying US-only inference via the

    i.j4i.i2
    inference_geo
    parameter incurs a 1.1x multiplier on all token pricing categories, including input tokens, output tokens, cache writes, and cache reads. Claude Mythos Preview, Opus 4.7, Opus 4.6, and Sonnet 4.6 include the full 1M token context window at standard pricing. Client-side tools are priced the same as any other Claude API request, while server-side tools may incur additional charges based on their sp…

  • [25] Claude Platform - Claude API Docsplatform.claude.com

    Opus 4.6 merekomendasikan pemikiran adaptif ( thinking: {type: "adaptive"} ); pemikiran manual ( type: "enabled" dengan budget_tokens ) sudah usang. Opus

  • [26] claude-code/CHANGELOG.md at main - GitHubplatform.claude.com

    Navigation Menu. # Search code, repositories, users, issues, pull requests... # Provide feedback. We read every piece of feedback, and take your input very seriously. # Saved searches. ## Use saved searches to filter your results more quickly. To see all available qualifiers, see our documentation. ## FilesExpand file tree. ## Footer. # CHANGELOG.md. ## Latest commit. ## History. ## Breadcrumbs. # CHANGELOG.md. ## File metadata and controls. # Changelog. ## Breadcrumbs.
    i.j4i.i2
    mktemp: No such file or directory
    .
    i.j4i.i2
    --exclude-dynamic-system-prompt-sections
    .
    i.j4i.i2
    403 "Authorization header is missing"
    .…

  • [27] Model deprecations - Claude API Docsplatform.claude.com

    On April 14, 2026, Anthropic notified developers using Claude Sonnet 4 and Claude Opus 4 models of their upcoming retirement on the Claude API. On February 19, 2026, Anthropic notified developers using Claude Haiku 3 model of its upcoming retirement on the Claude API. On December 19, 2025, Anthropic notified developers using Claude Haiku 3.5 model of its upcoming retirement on the Claude API. ### 2025-10-28: Claude Sonnet 3.7 model. On October 28, 2025, Anthropic notified developers using Claude Sonnet 3.7 model of its upcoming retirement on the Claude API. On August 13, 2025, Anthropic notif…

  • [28] Anthropic and Infosys collaborate to build AI agents for ...anthropic.com
    • Research. * News. Image 1: Anthropic and Infosys collaborate to build AI agents for telecommunications and other regulated industries. Anthropic and Infosys, a global leader in next-generation digital services and consulting founded and headquartered in Bengaluru, today announced a collaboration to develop and deliver enterprise AI solutions across telecommunications, financial ser…
  • [29] Introducing Claude 3.5 Sonnet - Anthropicanthropic.com
  • [30] Introducing Claude Opus 4.5 - Anthropicanthropic.com

    Image 1: Introducing Claude Opus 4.5. Image 2: Chart comparing frontier models on SWE-bench Verified where Opus 4.5 scores highest. If you’re a developer, simply use

    i.j4i.i2
    claude-opus-4-5-20251101
    via the Claude API. As we state in our [syst…

  • [31] Introducing Claude Opus 4.6 - Anthropicanthropic.com

    As we show in our extensive system card, Opus 4.6 also shows an overall safety profile as good as, or better than, any other frontier model in the industry, with low rates of misaligned behavior across safety evaluations. Image 2: Bar charts comparing Claude Opus 4.6 to other models on GDPval-AA. ![Image 3: Bar chart comparing Opus 4.6 to other models on Deep…

  • [32] Introducing Claude Sonnet 4.6 - Anthropicanthropic.com

    Introducing Claude Sonnet 4.6. For those on our Free and Pro plans, Claude Sonnet 4.6 is now the default model in claude.ai and Claude Cowork. As with every new Claude model, we’ve run extensive safety evaluations of Sonnet 4.6, which overall showed it to be as safe as, or safer than,…