Anthropic Claude release history.

Anthropic has shipped 23 major model releases and 27 other major launches since Claude arrived in March 2023. The latest model recorded here is Claude Opus 5.5, announced 22 Sep 2026. Every entry is dated from Anthropic’s own announcement and links to it.

50 MAJOR RELEASES2023–2026CHECKED 23 SEP 2026
NEWEST OPUS
Claude Opus 5.5 · 22 Sep 2026
NEWEST FABLE & MYTHOS
Claude Fable 5.1 and Mythos 5.1 · 1 Sep 2026
NEWEST SONNET
Claude Sonnet 5 · 30 Jun 2026
NEWEST HAIKU
Claude Haiku 4.5 · 15 Oct 2025
§ 01 — LINEAGE

How the Anthropic Claude models evolved.

MYTHOSProject Glasswing and Claude Mythos Preview — 2026-04-07PreviewClaude Fable 5 and Mythos 5 — 2026-06-095Claude Fable 5.1 and Mythos 5.1 — 2026-09-015.1FABLEClaude Fable 5 and Mythos 5 — 2026-06-095Claude Fable 5.1 and Mythos 5.1 — 2026-09-015.1OPUSClaude 3 family — 2024-03-043Claude 4 — 2025-05-224Claude Opus 4.1 — 2025-08-054.1Claude Opus 4.5 — 2025-11-244.5Claude Opus 4.6 — 2026-02-054.6Claude Opus 4.7 — 2026-04-164.7Claude Opus 4.8 — 2026-05-284.8Claude Opus 5 — 2026-07-245Claude Opus 5.5 — 2026-09-225.5SONNETClaude 3 family — 2024-03-043Claude 3.5 Sonnet — 2024-06-213.5Computer use and upgraded Claude 3.5 models — 2024-10-223.5 newClaude 3.7 Sonnet and Claude Code — 2025-02-243.7Claude 4 — 2025-05-224Claude Sonnet 4.5 — 2025-09-294.5Claude Sonnet 4.6 — 2026-02-174.6Claude Sonnet 5 — 2026-06-305HAIKUClaude 3 Haiku — 2024-03-133Computer use and upgraded Claude 3.5 models — 2024-10-223.5 newClaude Haiku 4.5 — 2025-10-154.5CLAUDE 1 & 2Claude launch — 2023-03-14Claude 1Claude 2 and claude.ai beta — 2023-07-112Claude 2.1 — 2023-11-212.1
CURRENTSUPERSEDEDRETIREDLARGER SQUARE = WATERSHED
§ 02 — MODELS

Every Anthropic Claude model release.

23 model releases, newest first. “Current” is the newest release in its line; older ones are superseded, and retired ones have left Anthropic’s apps or API. Where a model is in our benchmark tables, the last column links to its scores, price in AUD and Australian availability.

MODEL RELEASES23 RELEASES
Claude model releases, newest first
RELEASEANNOUNCEDLINESTATUSBENCHMARKS
Claude Opus 5.5OpusCURRENTOpus 5.5 →
Claude Fable 5.1 and Mythos 5.1Fable · MythosCURRENTFable 5.1 →
Claude Opus 5OpusSUPERSEDEDOpus 5 →
Claude Sonnet 5SonnetCURRENTSonnet 5 →
Claude Fable 5 and Mythos 5WATERSHEDFable · MythosSUPERSEDEDFable 5 →
Claude Opus 4.8OpusSUPERSEDEDOpus 4.8 →
Claude Opus 4.7OpusSUPERSEDEDOpus 4.7 →
Project Glasswing and Claude Mythos PreviewWATERSHEDMythosSUPERSEDED
Claude Sonnet 4.6SonnetSUPERSEDEDSonnet 4.6 →
Claude Opus 4.6OpusSUPERSEDEDOpus 4.6 →
Claude Opus 4.5OpusSUPERSEDED
Claude Haiku 4.5HaikuCURRENT
Claude Sonnet 4.5SonnetSUPERSEDED
Claude Opus 4.1OpusRETIRED
Claude 4WATERSHEDOpus · SonnetRETIRED
Claude 3.7 Sonnet and Claude CodeWATERSHEDSonnetSUPERSEDED
Computer use and upgraded Claude 3.5 modelsWATERSHEDSonnet · HaikuSUPERSEDED
Claude 3.5 SonnetWATERSHEDSonnetSUPERSEDED
Claude 3 HaikuHaikuSUPERSEDED
Claude 3 familyWATERSHEDOpus · SonnetSUPERSEDED
Claude 2.1Claude 1 & 2SUPERSEDED
Claude 2 and claude.ai betaWATERSHEDClaude 1 & 2SUPERSEDED
Claude launchWATERSHEDClaude 1 & 2SUPERSEDED
§ 03 — TIMELINE

Every major release, year by year.

50 OF 50

202619 RELEASES

  1. MODEL

    Claude Opus 5.5

    • GENERALLY AVAILABLE
    • CURRENT
    • CLAUDE 5

    Claude Opus 5.5 launched as the first model in the Claude 5.5 family, at $4/$20 per million input/output tokens (Opus 5 was $5/$25), with a 1M token context window and always-on adaptive thinking. Anthropic said Sonnet 5.5 and Haiku 5.5 would follow in the coming weeks.

    Why it mattered. Anthropic said it performs at about the level of Fable 5.1 on most work while costing around 40% less than Opus 5 on typical workloads, and it came with higher five-hour usage limits on paid plans.

    AVAILABLE TO
    Pro · Max · Team · Enterprise · API

  2. PRODUCT

    Cowork and chat merged into one Claude

    • LAUNCHED
    • CLAUDE 5

    Anthropic merged Claude Cowork and chat into a single Claude interface and launched Claude Docs and Claude Slides, with Claude Design also usable inside conversations. The merge is rolling out to Pro and Max first, with Team and Free to follow and Enterprise admins given at least 30 days' notice. Docs, Slides and Design are in beta on paid plans.

    Why it mattered. Users no longer have to choose between chat and Cowork, because one conversation can produce documents, slides and designs using the same context, skills and connectors.

    AVAILABLE TO
    Free · Pro · Max · Team · Enterprise

  3. MODEL

    Claude Fable 5.1 and Mythos 5.1

    • GENERALLY AVAILABLE
    • CURRENT
    • CLAUDE 5

    Claude Fable 5.1 succeeded Fable 5 at the same $10/$50 per million token price, with cache reads cut to $0.25 per million. Claude Mythos 5.1 is the same model with lighter safeguards and is available only through Anthropic's trusted access programs.

    Why it mattered. It improved coding, knowledge work and scientific research at the top tier while lowering the cost of long agentic runs through cheaper cache reads.

    AVAILABLE TO
    Pro · Max · Team · Enterprise · API

  4. MODEL

    Claude Opus 5

    • GENERALLY AVAILABLE
    • SUPERSEDED
    • CLAUDE 5

    Claude Opus 5 launched at $5/$25 per million input/output tokens (the same as Opus 4.8), with a 1M token context window, 128k max output and an optional fast mode. It became the default model on Max and the strongest model on Pro.

    Why it mattered. Anthropic reported that at max effort it came within 0.5% of Fable 5's peak score at about half the cost per task, so it offered near-frontier capability at Opus pricing.

    AVAILABLE TO
    Pro · Max · API

  5. MODEL

    Claude Sonnet 5

    • GENERALLY AVAILABLE
    • CURRENT
    • CLAUDE 5

    Claude Sonnet 5 launched as the default model for Free and Pro users and on the API at $2/$10 per million input/output tokens, with a 1M token context window and 128k max output. On 10 August Anthropic made the introductory price permanent.

    Why it mattered. It brought close to Opus 4.8-level agentic coding and tool use to the mid-priced tier and to the Free plan, at a lower API price than Sonnet 4.6.

    AVAILABLE TO
    Free · Pro · Max · Team · Enterprise · API

  6. PRODUCT

    Claude Science workbench

    • BETA
    • CLAUDE 5

    Anthropic released Claude Science, a desktop research workbench with more than 60 curated skills and connectors for fields such as genomics, proteomics and cheminformatics, where every result traces back to its code. It launched in beta for Pro, Max, Team and Enterprise.

    Why it mattered. It was Anthropic's first dedicated app for scientific research, giving researchers reproducible analysis in one place instead of a general chat interface.

    AVAILABLE TO
    Pro · Max · Team · Enterprise

  7. PRODUCT

    Claude Tag in Slack

    • BETA
    • CLAUDE 5

    Anthropic launched Claude Tag, which lets team members tag @Claude in Slack channels and hand it tasks that it works on asynchronously, using the tools, data and channel context an admin has granted it. It launched in beta for Team and Enterprise and replaced the existing Claude in Slack app.

    Why it mattered. It moved Claude from a chat window into a shared team workspace as a delegated co-worker, which changes how organisations roll it out and govern it.

    AVAILABLE TO
    Team · Enterprise

  8. MODEL

    Claude Fable 5 and Mythos 5

    • WATERSHED
    • LAUNCHED
    • SUPERSEDED
    • CLAUDE 5

    Anthropic launches Claude Fable 5 for general use with safeguards and Claude Mythos 5 for restricted trusted-access use. Three days later, on 12 June, access to both is suspended after the US government applies export controls; Fable 5 returns worldwide on 1 July with an improved safety classifier, and Mythos 5 is restored for a set of US organisations.

    Why it mattered. Fable 5 marked a new top-end Claude capability tier. The suspension was the first time a released Claude model was withdrawn from every user for regulatory reasons, a reminder that access to frontier models can change at short notice.

    AVAILABLE TO
    Claude users · Developers · Trusted-access partners

  9. TRUST & SAFETY

    Claude Security and Glasswing expansion

    • EXPANDED
    • COWORK AND MYTHOS

    Anthropic expands Project Glasswing to about 150 additional organisations and says it recently released Claude Security for scanning codebases and suggesting patches with public frontier models.

    Why it mattered. This moved Claude's cyber-defence work from a limited Mythos Preview programme toward a broader product and partner-access model for defensive security workflows.

    AVAILABLE TO
    Security teams · Critical infrastructure · Developers

  10. MODEL

    Claude Opus 4.8

    • LAUNCHED
    • SUPERSEDED
    • COWORK AND MYTHOS

    Anthropic upgrades Claude Opus to 4.8, with improvements over Opus 4.7 in benchmarks, coding, agentic skills, reasoning, and practical knowledge work.

    Why it mattered. Opus 4.8 kept Claude's flagship model moving quickly after Opus 4.7, while adding effort control and cheaper fast mode for higher-throughput use.

    AVAILABLE TO
    Claude users · Developers · Claude Code

  11. PRODUCT

    Claude Design

    • RESEARCH PREVIEW
    • COWORK AND MYTHOS

    Anthropic launches Claude Design, an Anthropic Labs product for collaborating with Claude on designs, prototypes, slides, and one-pagers.

    Why it mattered. Claude Design carved out a specialised creative-work surface powered by Opus 4.7's stronger vision and professional output quality.

    AVAILABLE TO
    Pro · Max · Team · Enterprise

  12. MODEL

    Claude Opus 4.7

    • LAUNCHED
    • SUPERSEDED
    • COWORK AND MYTHOS

    Anthropic launches Claude Opus 4.7 with gains in advanced software engineering, complex multi-step tasks, higher-resolution vision, professional work, and cyber safeguards.

    Why it mattered. Opus 4.7 gave Claude a generally available model with stronger agent reliability while Anthropic kept Mythos Preview gated.

    AVAILABLE TO
    Claude users · Developers · Cloud customers

  13. AGENT

    Claude Cowork generally available

    • GA
    • COWORK AND MYTHOS

    Claude Cowork becomes generally available on macOS and Windows through Claude Desktop, with analytics, OpenTelemetry, and Enterprise role-based access controls.

    Why it mattered. GA made Cowork a real product surface for delegating knowledge work to Claude rather than an experimental preview.

    AVAILABLE TO
    Claude Desktop · Team · Enterprise

  14. TRUST & SAFETY

    Project Glasswing and Claude Mythos Preview

    • WATERSHED
    • RESEARCH PREVIEW
    • SUPERSEDED
    • COWORK AND MYTHOS

    Anthropic announces Project Glasswing, a critical-software security initiative powered by Claude Mythos Preview, an unreleased frontier model for defensive cybersecurity partners.

    Why it mattered. Glasswing showed a new class of frontier capability and a cautious deployment model for high-risk cyber work.

    AVAILABLE TO
    Project partners · Security teams

  15. AGENT

    Computer use in Cowork and Claude Code

    • RESEARCH PREVIEW
    • COWORK AND MYTHOS

    Pro and Max users can give Claude access to computer use in Cowork and Claude Code, letting it open files, run tools, click, type, and navigate.

    Why it mattered. This was the practical consumer/workstation version of computer use, bringing Claude's agent capabilities onto local machines.

    AVAILABLE TO
    Pro · Max

  16. MODEL

    Claude Sonnet 4.6

    • LAUNCHED
    • SUPERSEDED
    • COWORK AND MYTHOS

    Anthropic launches Claude Sonnet 4.6 with improvements across coding, computer use, long-context reasoning, agent planning, knowledge work, design, and a 1M context beta.

    Why it mattered. Sonnet 4.6 made stronger agentic and professional performance available as the default Claude model for Free and Pro users.

    AVAILABLE TO
    Free · Pro · Developers · Cowork users

  17. PRODUCT

    Claude for PowerPoint and Excel upgrades

    • LAUNCHED
    • COWORK AND MYTHOS

    Claude for PowerPoint launches, Claude for Excel upgrades to Opus 4.6 with native Excel operations, and health data support expands on mobile.

    Why it mattered. The update made Claude more directly useful inside core knowledge-work applications rather than only beside them.

    AVAILABLE TO
    Pro · Max · Team · Enterprise

  18. MODEL

    Claude Opus 4.6

    • LAUNCHED
    • SUPERSEDED
    • COWORK AND MYTHOS

    Anthropic launches Claude Opus 4.6 with stronger coding, longer agentic task reliability, better debugging, everyday work gains, and a 1M token context window in beta.

    Why it mattered. Opus 4.6 pushed Claude's smartest model toward long-horizon work across code, documents, spreadsheets, presentations, and Cowork.

    AVAILABLE TO
    Claude users · Developers · Cloud customers

  19. AGENT

    Claude Cowork research preview

    • WATERSHED
    • RESEARCH PREVIEW
    • COWORK AND MYTHOS

    Anthropic launches Cowork research preview on Claude Desktop for Max users, bringing Claude Code-style agentic capabilities to local knowledge work.

    Why it mattered. Cowork was the bridge from coding agents to general professional agents that can work with local files and MCP integrations.

    AVAILABLE TO
    Max

202515 RELEASES

  1. PRODUCT

    Claude for Excel beta and context compaction

    • BETA
    • CLAUDE 4 AGENTS

    Claude for Excel enters beta for Max, Team, and Enterprise users, while context compaction reduces length-limit errors in long conversations.

    Why it mattered. Excel and compaction pushed Claude toward durable professional workflows with long-running context and direct deliverable editing.

    AVAILABLE TO
    Max · Team · Enterprise

  2. MODEL

    Claude Opus 4.5

    • LAUNCHED
    • SUPERSEDED
    • CLAUDE 4 AGENTS

    Anthropic launches Claude Opus 4.5 across apps, API, and major cloud platforms, with improvements in coding, agents, computer use, research, slides, and spreadsheets.

    Why it mattered. Opus 4.5 was a major frontier-model release that pushed Claude into deeper professional work and long-running agent tasks.

    AVAILABLE TO
    Pro · Max · Team · Enterprise · Developers

  3. PLAN

    Claude in Microsoft Foundry and Microsoft 365 Copilot

    • PREVIEW
    • CLAUDE 4 AGENTS

    Anthropic expands its Microsoft partnership, bringing Claude models to Foundry and Claude-powered options into Microsoft 365 Copilot experiences.

    Why it mattered. This gave Claude another major cloud and enterprise distribution channel alongside its own API, Bedrock, and Vertex AI.

    AVAILABLE TO
    Azure customers · Microsoft 365 users

  4. AGENT

    Agent Skills

    • BETA
    • CLAUDE 4 AGENTS

    Anthropic launches Agent Skills in beta, letting Claude dynamically load folders of instructions, scripts, and resources for specialised tasks.

    Why it mattered. Skills created a portable way to teach agents repeatable workflows across documents, spreadsheets, PDFs, and custom domains.

    AVAILABLE TO
    Developers · Organizations

  5. MODEL

    Claude Haiku 4.5

    • LAUNCHED
    • CURRENT
    • CLAUDE 4 AGENTS

    Anthropic releases Claude Haiku 4.5, a fast model matching Sonnet 4-level coding and computer-use performance at lower cost.

    Why it mattered. Haiku 4.5 made agentic and coding capabilities more responsive for high-volume and latency-sensitive use cases.

    AVAILABLE TO
    Claude users · Developers

  6. MODEL

    Claude Sonnet 4.5

    • LAUNCHED
    • SUPERSEDED
    • CLAUDE 4 AGENTS

    Anthropic releases Claude Sonnet 4.5 with major gains in coding, agents, computer use, reasoning, and math.

    Why it mattered. Sonnet 4.5 re-centered Claude around practical agentic work and became the default engine for many browser and coding workflows.

    AVAILABLE TO
    Claude users · Developers

  7. PRODUCT

    Memory for Team and incognito chats

    • ROLLING OUT
    • CLAUDE 4 AGENTS

    Claude adds memory for Team users and incognito chats for Free, Pro, Max, and Team plans.

    Why it mattered. Memory made Claude more personal and durable, while incognito controls gave users a way to keep sensitive chats out of memory.

    AVAILABLE TO
    Free · Pro · Max · Team

  8. AGENT

    Claude in Chrome

    • EXPERIMENTAL
    • CLAUDE 4 AGENTS

    Anthropic introduces an experimental Chrome extension that lets Claude read, click, and navigate websites alongside users.

    Why it mattered. Claude in Chrome extended computer use from API demos toward a consumer-facing browser assistant.

    AVAILABLE TO
    Max waitlist

  9. MODEL

    1M context beta for Claude Sonnet 4

    • BETA
    • CLAUDE 4 AGENTS

    Anthropic launches beta support for a 1M token context window in Claude Sonnet 4 on the Claude API and Amazon Bedrock.

    Why it mattered. The 1M beta extended Claude's long-context identity into larger codebases, research corpora, and enterprise archives.

    AVAILABLE TO
    API organizations · Amazon Bedrock

  10. MODEL

    Claude Opus 4.1

    • LAUNCHED
    • RETIRED
    • CLAUDE 4 AGENTS

    Anthropic releases Claude Opus 4.1 with improvements to agentic tasks, real-world coding, reasoning, research, and analysis.

    Why it mattered. Opus 4.1 kept Claude's top model competitive while larger model jumps were still ahead.

    5 Aug 2026: Retired from the Claude API (deprecation announced 5 June 2026).

    AVAILABLE TO
    Paid Claude · Claude Code · Developers

  11. MODEL

    Claude 4

    • WATERSHED
    • LAUNCHED
    • RETIRED
    • CLAUDE 4 AGENTS

    Anthropic introduces Claude Opus 4 and Claude Sonnet 4, plus Claude Code GA, extended thinking with tools, and new API capabilities for agents.

    Why it mattered. Claude 4 marked Anthropic's agentic turn, pairing stronger hybrid models with Claude Code and tools for long-running work.

    15 Jun 2026: Claude Opus 4 and Sonnet 4 were retired from the Claude API; Anthropic recommends Opus 4.8 and Sonnet 4.6.

    AVAILABLE TO
    Free · Pro · Max · Team · Enterprise · Developers

  12. TRUST & SAFETY

    ASL-3 protections for Claude Opus 4

    • ACTIVATED
    • CLAUDE 4 AGENTS

    Anthropic activates AI Safety Level 3 deployment and security protections in conjunction with Claude Opus 4.

    Why it mattered. The launch showed Anthropic's Responsible Scaling Policy becoming an operational part of frontier model deployment.

    AVAILABLE TO
    Claude users · Developers

  13. AGENT

    Integrations, advanced Research, and global web search

    • LAUNCHED
    • CLAUDE 4 AGENTS

    Anthropic introduces Integrations for remote MCP servers, expands Research to search Google Workspace and connected apps, and makes web search global for paid users.

    Why it mattered. Claude became a more connected research agent, able to combine web, workspace, and app context into longer source-backed reports.

    AVAILABLE TO
    Paid plans · Developers

  14. PLAN

    Claude Max

    • LAUNCHED
    • COMPUTER-USE PLATFORM

    Anthropic introduces Max, a plan for users who collaborate with Claude extensively and need higher limits for demanding projects.

    Why it mattered. Max created a consumer and professional tier for heavy Claude use, similar to a frontier-compute plan for individuals.

    AVAILABLE TO
    Max

  15. AGENT

    Claude 3.7 Sonnet and Claude Code

    • WATERSHED
    • RESEARCH PREVIEW
    • SUPERSEDED
    • COMPUTER-USE PLATFORM

    Anthropic launches Claude 3.7 Sonnet, its first hybrid reasoning model, and introduces Claude Code as a terminal coding agent research preview.

    Why it mattered. This was the first clear bridge between Claude's reasoning model story and hands-on agentic software development.

    AVAILABLE TO
    Free · Pro · Team · Enterprise · Developers

202411 RELEASES

  1. PLATFORM

    Model Context Protocol

    • WATERSHED
    • OPEN SOURCED
    • COMPUTER-USE PLATFORM

    Anthropic introduces MCP as an open standard for connecting AI assistants to external tools, repositories, and business systems.

    Why it mattered. MCP became a foundational piece of the agent ecosystem, giving Claude and other assistants a common way to reach real data and tools.

    AVAILABLE TO
    Developers · Enterprises

  2. AGENT

    Computer use and upgraded Claude 3.5 models

    • WATERSHED
    • PUBLIC BETA
    • SUPERSEDED
    • COMPUTER-USE PLATFORM

    Anthropic releases an upgraded Claude 3.5 Sonnet, announces Claude 3.5 Haiku, and introduces computer use in public beta.

    Why it mattered. Computer use was a watershed agent moment: Claude could perceive screens and take actions through ordinary software interfaces.

    AVAILABLE TO
    Developers · Claude users

  3. PLAN

    Claude Enterprise

    • LAUNCHED
    • CLAUDE 3 WORKSPACE

    Anthropic announces the Claude Enterprise plan with a 500K context window, more usage, native GitHub integration, SSO, role permissions, and admin tools.

    Why it mattered. Enterprise positioned Claude as sanctioned company infrastructure for knowledge work across large organisations.

    AVAILABLE TO
    Enterprise

  4. PRODUCT

    Artifacts generally available

    • GA
    • CLAUDE 3 WORKSPACE

    Artifacts become available for Free, Pro, and Team users, including creation and viewing on iOS and Android.

    Why it mattered. Artifacts made Claude feel less like a chat box and more like a tool for building, visualising, and iterating on work.

    AVAILABLE TO
    Free · Pro · Team

  5. PRODUCT

    Projects in Claude

    • LAUNCHED
    • CLAUDE 3 WORKSPACE

    Projects become available on Claude.ai for Pro and Team customers, letting users group chats, knowledge, and work context.

    Why it mattered. Projects made Claude more durable for real work by organising ongoing tasks instead of treating every chat as standalone.

    AVAILABLE TO
    Pro · Team

  6. MODEL

    Claude 3.5 Sonnet

    • WATERSHED
    • LAUNCHED
    • SUPERSEDED
    • CLAUDE 3 WORKSPACE

    Anthropic launches Claude 3.5 Sonnet, outperforming Claude 3 Opus on many evaluations at Sonnet speed and cost, alongside the Artifacts preview.

    Why it mattered. Sonnet 3.5 became the defining everyday Claude model and introduced a visible work surface for generated apps, documents, and designs.

    AVAILABLE TO
    Free · Pro · Team · Developers

  7. PLATFORM

    Tool use generally available

    • GA
    • CLAUDE 3 WORKSPACE

    Tool use becomes generally available across the Claude API, Amazon Bedrock, and Google Cloud's Vertex AI.

    Why it mattered. Tool use was the platform primitive that let Claude move from answering questions to orchestrating external systems.

    AVAILABLE TO
    Developers · Cloud customers

  8. PLAN

    Claude Team plan

    • LAUNCHED
    • CLAUDE 3 WORKSPACE

    Anthropic launches a Team plan with higher usage, workspace controls, user management, and centralised billing.

    Why it mattered. Team filled the gap between individual Pro accounts and enterprise sales-led deployments.

    AVAILABLE TO
    Team

  9. PRODUCT

    Claude for iOS

    • LAUNCHED
    • CLAUDE 3 WORKSPACE

    Anthropic releases the Claude iOS app with synced chat history and photo upload support.

    Why it mattered. Mobile made Claude easier to use in daily workflows and gave the assistant a persistent consumer surface.

    AVAILABLE TO
    Free · Pro · Team

  10. MODEL

    Claude 3 Haiku

    • LAUNCHED
    • SUPERSEDED
    • CLAUDE 3 WORKSPACE

    Claude 3 Haiku launches as Anthropic's fastest and most affordable model in the Claude 3 generation.

    Why it mattered. Haiku made Claude 3 practical for high-volume enterprise tasks that needed speed, vision, and lower cost.

    AVAILABLE TO
    Pro · Developers · Enterprise

  11. MODEL

    Claude 3 family

    • WATERSHED
    • LAUNCHED
    • SUPERSEDED
    • CLAUDE 3 WORKSPACE

    Anthropic announces Claude 3 Opus, Sonnet, and Haiku with multimodal capabilities and stronger benchmarks across reasoning, coding, and vision.

    Why it mattered. Claude 3 turned the product into a tiered model family and made Opus a flagship frontier competitor.

    AVAILABLE TO
    Free · Pro · Developers · Businesses

20235 RELEASES

  1. MODEL

    Claude 2.1

    • LAUNCHED
    • SUPERSEDED
    • ASSISTANT FOUNDATION

    Claude 2.1 adds a 200K token context window, lower hallucination rates, system prompts, and beta tool use.

    Why it mattered. The 200K context window and early tool use made Claude more credible for long-document enterprise workflows.

    AVAILABLE TO
    Claude.ai · Developers

  2. PLAN

    Claude Pro

    • LAUNCHED
    • ASSISTANT FOUNDATION

    Anthropic introduces Claude Pro for the Claude.ai chat experience, offering more usage, priority access, and early feature access.

    Why it mattered. Pro created a durable consumer business model and a release lane for heavier Claude users.

    AVAILABLE TO
    Pro

  3. MODEL

    Claude 2 and claude.ai beta

    • WATERSHED
    • BETA
    • SUPERSEDED
    • ASSISTANT FOUNDATION

    Claude 2 launches with better coding, math, reasoning, longer responses, API access, and a public beta chat site at claude.ai.

    Why it mattered. Claude moved from mostly partner/API distribution to a direct consumer-facing assistant experience.

    AVAILABLE TO
    Free beta · Developers

  4. MODEL

    100K context windows

    • LAUNCHED
    • ASSISTANT FOUNDATION

    Anthropic expands Claude's context window from 9K to 100K tokens for long documents and multi-hour conversations.

    Why it mattered. Long context became one of Claude's signature advantages and helped define the product for research, legal, and enterprise work.

    AVAILABLE TO
    Developers · Businesses

  5. PRODUCT

    Claude launch

    • WATERSHED
    • LAUNCHED
    • SUPERSEDED
    • ASSISTANT FOUNDATION

    Anthropic introduces Claude and Claude Instant after closed alpha testing with partners including Notion, Quora, and DuckDuckGo.

    Why it mattered. This established Claude as Anthropic's helpful, honest, and harmless assistant family across chat and API use cases.

    AVAILABLE TO
    Partners · Developers
§ 04 — ERAS

6 eras of Anthropic Claude.

  1. ERA 01MARCH 2023 – NOVEMBER 2023

    Assistant foundation

    Claude goes from partner access to a public assistant: 100K context, Claude 2 and claude.ai, the Pro plan and 200K context with Claude 2.1.

    5 releases, from Claude launch →
  2. ERA 02MARCH 2024 – SEPTEMBER 2024

    Claude 3 workspace

    The Claude 3 family, Claude 3.5 Sonnet, Projects, Artifacts, the iPhone app and the Team and Enterprise plans turn Claude into a work product.

    9 releases, from Claude 3 family →
  3. ERA 03OCTOBER 2024 – APRIL 2025

    Computer-use platform

    Computer use, the Model Context Protocol, Claude 3.7 Sonnet and Claude Code make Claude a platform for agents that act, not just answer.

    4 releases, from Computer use and upgraded Claude 3.5 models →
  4. ERA 04MAY 2025 – NOVEMBER 2025

    Claude 4 agents

    Claude 4, Opus 4.1, Sonnet and Haiku 4.5, Opus 4.5, Research, Claude in Chrome, Skills and memory build Claude into an agent platform for work.

    13 releases, from Integrations, advanced Research, and global web search →
  5. ERA 05JANUARY 2026 – JUNE 2026

    Cowork and Mythos

    Cowork brings agents to everyday office work, Opus 4.6 to 4.8 arrive every few months, and Mythos opens a restricted frontier tier for security work.

    11 releases, from Claude Cowork research preview →
  6. ERA 06JUNE 2026 – NOW

    Claude 5

    Fable 5, Sonnet 5, Opus 5, Fable 5.1 and Opus 5.5 ship within four months, a short export-control suspension tests access, and chat and Cowork merge into one Claude.

    8 releases, from Claude Fable 5 and Mythos 5 →
§ 05 — QUESTIONS

Anthropic Claude releases, answered.

As of our 23 Sep 2026 review, Claude Opus 5.5, announced by Anthropic on 22 Sep 2026. Claude Opus 5.5 launched as the first model in the Claude 5.5 family, at $4/$20 per million input/output tokens (Opus 5 was $5/$25), with a 1M token context window and always-on adaptive thinking. Anthropic said Sonnet 5.5 and Haiku 5.5 would follow in the coming weeks.

14 Mar 2023: Claude launch. Anthropic introduces Claude and Claude Instant after closed alpha testing with partners including Notion, Quora, and DuckDuckGo.

As of 23 Sep 2026, the newest release in each line is: Claude Haiku 4.5 (15 Oct 2025), Claude Sonnet 5 (30 Jun 2026), Claude Fable 5.1 and Mythos 5.1 (1 Sep 2026) and Claude Opus 5.5 (22 Sep 2026). Older models in those lines have been superseded, and some have been retired from Anthropic’s apps.

50 on this timeline: 23 model releases and 27 product, platform, agent and plan launches. It leaves out incremental updates, settings and feature drops — only releases that changed what Claude could do or who could use it.

Claude Sonnet 5: it can run with requests kept in Australia on AWS Bedrock in Melbourne. Claude Fable 5.1: it can be called from AWS Bedrock in Sydney and Melbourne and Google Vertex AI in Sydney, but only through global routing, so requests may be processed outside Australia. Claude Opus 5.5: it can run with requests kept in Australia on AWS Bedrock in Sydney and Melbourne. Our benchmark pages track Australian cloud availability for each model weekly.

Dates are the day Anthropic announced each release, taken from Anthropic’s own site and documentation.

Major releases only: new models, products, plans and capability launches. Retirements are noted on the model they retired.

Checked 23 Sep 2026 · How the current models compare →

MAKE SENSE OF THE NEXT RELEASE

Choose AI improvements around your business needs.

DISCUSS YOUR AI OPTIONSAI SUPPORT & MAINTENANCE