Anthropic Claude Models — Complete Guide
🟣 Anthropic Claude Models — Complete Guide
Section titled “🟣 Anthropic Claude Models — Complete Guide”Anthropic organises its Claude models into named tiers that have remained consistent across every generation: Opus (maximum capability), Sonnet (balanced), and Haiku (fast and cost-efficient) — joined in the Claude 5 generation by Fable, a new tier positioned above Opus. Each generation raises the bar across all tiers.
🗺️ The Claude Generation Map
Section titled “🗺️ The Claude Generation Map”| 📅 Generation | 🔮 Opus | ⚖️ Sonnet | ⚡ Haiku | 🎭 Fable | 🗓️ Year |
|---|---|---|---|---|---|
| Claude 3 | Claude 3 Opus | Claude 3 Sonnet | Claude 3 Haiku | — | Mar 2024 |
| Claude 3.5 | — | Claude 3.5 Sonnet (Jun + Oct 2024) | Claude 3.5 Haiku | — | Oct 2024 |
| Claude 3.7 | — | Claude 3.7 Sonnet | — | — | Feb 2025 |
| Claude 4 | Claude Opus 4 | Claude Sonnet 4 | — | — | May 2025 |
| Claude 4.5 | Claude Opus 4.5 | Claude Sonnet 4.5 | Claude Haiku 4.5 | — | Sep–Nov 2025 |
| Claude 4.6 | Claude Opus 4.6 | Claude Sonnet 4.6 | — | — | Feb 2026 |
| Claude 4.7 | Claude Opus 4.7 | — | — | — | Apr 2026 |
| Claude 4.8 | Claude Opus 4.8 | — | — | — | May 2026 |
| Claude 5 | Claude Opus 5 | Claude Sonnet 5 | Claude Haiku 4.5 (carried over) | Claude Fable 5 ✅ Current top tier | Jun 2026– |
📊 Current Active Models (2026)
Section titled “📊 Current Active Models (2026)”These are the models in active use and recommended for new projects:
| 🤖 Model | 🏷️ Tier | 📅 Released | 🎯 Best For |
|---|---|---|---|
| 🎭 Claude Fable 5 | Top tier | 2026 | Anthropic’s most capable widely-released model — hardest reasoning and agentic tasks |
| 🔮 Claude Opus 5 | Flagship | 2026 | Complex reasoning, long-horizon agents |
| ⚖️ Claude Sonnet 5 | Balanced | Jun 2026 | Day-to-day coding, writing, production |
| ⚡ Claude Haiku 4.5 | Fast | Oct 2025 | Classification, routing, high-volume tasks — still current, no Haiku 5 yet |
Claude Opus 4.8 and Claude Sonnet 4.6 remain available via the API but are superseded by the Claude 5 generation for new projects.
🎭 Fable Tier — New Top Tier (Claude 5)
Section titled “🎭 Fable Tier — New Top Tier (Claude 5)”The Fable tier debuted with the Claude 5 generation as a new level above Opus — Anthropic’s most capable widely-released model, with always-on adaptive thinking and a very large context window.
🎭 Claude Fable 5 — Anthropic's Most Capable Model
| 🔧 Property | 📋 Detail |
|---|---|
| 🏷️ Tier | New — sits above Opus |
| 🎯 Focus | State-of-the-art results across nearly all published benchmarks |
| 🌐 Access | Anthropic API · Claude.ai (availability may be limited/staged) |
✅ Ideal for:
- 🧩 The hardest reasoning and agentic tasks, where Opus-tier quality isn’t quite enough
- 🤖 Long-horizon, many-step agentic workflows
Check Anthropic’s official model docs for current pricing and exact specifications — this tier is new enough that details are still evolving.
🔮 Opus Tier — Maximum Capability
Section titled “🔮 Opus Tier — Maximum Capability”The Opus tier remains Anthropic’s flagship below Fable — strong reasoning, deep coding ability, and long context handling.
🔮 Claude Opus 5 — Current Opus Generation
| 🔧 Property | 📋 Detail |
|---|---|
| 🏷️ Generation | Claude 5 |
| 🎯 Focus | Complex reasoning · long-horizon agentic coding · professional work |
| 🌐 Access | Anthropic API · Claude.ai |
✅ Ideal for:
- 🚨 Production incident debugging where failure is expensive
- 🏗️ Architecture decisions and migration planning
- 📁 Large cross-file refactors requiring coherent multi-step plans
- 👀 Reviewing high-stakes pull requests for security and correctness
- 🤖 Long-horizon agentic tasks that span many tool calls and steps
- 📄 Analysing very long documents (contracts, research papers, codebases)
💡 Example:
“Review this entire microservices codebase (attached). Identify the top three architectural risks, propose a phased remediation plan, and highlight which services must be touched first to unblock the others.”
📅 Earlier Opus Models
| 🤖 Model | 📅 Released | 📋 Notes |
|---|---|---|
| Claude Opus 4.8 | May 2026 | Previous flagship, superseded by Opus 5 |
| Claude Opus 4.7 | Apr 2026 | Incremental improvement over 4.6 |
| Claude Opus 4.6 | Feb 2026 | Preferred by independent raters over comparable models |
| Claude Opus 4.5 | Nov 2025 | First Opus with significantly expanded agentic ability |
| Claude Opus 4 | May 2025 | First of the Claude 4 generation |
| Claude 3 Opus | Mar 2024 | Original flagship — strong reasoning, 200K context |
Earlier Opus versions are still accessible via the API but newer generations are recommended for new projects.
⚖️ Sonnet Tier — Balanced Default
Section titled “⚖️ Sonnet Tier — Balanced Default”The Sonnet tier is the recommended default for most production use cases — strong enough for hard tasks, fast and cost-efficient enough for everyday work.
⚖️ Claude Sonnet 5 — Current Balanced Model (Jun 2026)
| 🔧 Property | 📋 Detail |
|---|---|
| 📅 Released | June 30, 2026 |
| 🏷️ Generation | Claude 5 |
| 🎯 Focus | Coding · writing · analysis · production assistant workflows |
| 🌐 Access | Anthropic API · Claude.ai · Amazon Bedrock · Google Cloud Vertex AI |
✅ Ideal for:
- 💻 Day-to-day coding: adding features, fixing bugs, writing tests
- 📝 Writing and editing: technical docs, summaries, emails
- 🔍 Code review and explanation
- 🤖 Building production AI assistants and agents
- 📊 Data analysis and structured extraction
💡 Example:
“Add input validation and rate limiting to this Express.js API endpoint. Write Jest tests for the happy path and three error cases. Update the OpenAPI spec.”
📅 Earlier Sonnet Models
| 🤖 Model | 📅 Released | 📋 Notable Feature |
|---|---|---|
| Claude Sonnet 4.6 | Feb 2026 | Previous balanced default, superseded by Sonnet 5 |
| Claude Sonnet 4.5 | Sep 2025 | Improved agentic performance |
| Claude Sonnet 4 | May 2025 | First Claude 4 generation Sonnet |
| Claude 3.7 Sonnet | Feb 2025 | Introduced extended thinking / hybrid reasoning |
| Claude 3.5 Sonnet v2 | Oct 2024 | Major upgrade — computer use capability |
| Claude 3.5 Sonnet | Jun 2024 | Landmark release — outperformed Opus 3 at Sonnet price |
| Claude 3 Sonnet | Mar 2024 | First Sonnet, balanced speed and quality |
📌 Notable milestone: Claude 3.5 Sonnet (June 2024) was the first Anthropic model to outperform the previous generation’s Opus tier — a landmark in the efficiency progression.
⚡ Haiku Tier — Fast and Cost-Efficient
Section titled “⚡ Haiku Tier — Fast and Cost-Efficient”The Haiku tier is designed for tasks where speed and cost matter more than maximum reasoning depth — classification, routing, extraction, and simple transformations at scale.
⚡ Claude Haiku 4.5 — Current Fast Model (Oct 2025)
| 🔧 Property | 📋 Detail |
|---|---|
| 📅 Released | October 2025 |
| 📄 Context window | 200,000 tokens |
| 🎯 Focus | Speed · cost efficiency · high-volume tasks |
| 🌐 Access | Anthropic API · Claude.ai · Amazon Bedrock |
✅ Ideal for:
- 🏷️ Classifying support tickets, emails, or reviews at scale
- 🔀 Routing user requests to the right handler
- 🔎 Extracting structured fields from short text
- ✏️ Rewriting short content (subject lines, summaries)
- 📱 Latency-sensitive applications
💡 Example:
“Classify each of these 10,000 support tickets as: billing, login, technical, or feedback. Return a JSON array with ticket ID and label.”
📅 Earlier Haiku Models
| 🤖 Model | 📅 Released | 📋 Notes |
|---|---|---|
| Claude 3.5 Haiku | Nov 2024 | First 3.5 Haiku — faster than 3 Haiku with better quality |
| Claude 3 Haiku | Mar 2024 | Original Haiku — lightest and fastest Claude 3 |
| Claude Instant | Mar 2023 | Pre-Claude 3 era fast model |
🧠 Key Claude Capabilities (All Current Models)
Section titled “🧠 Key Claude Capabilities (All Current Models)”| ✅ Capability | 📋 Detail |
|---|---|
| 👁️ Vision | Understands images, screenshots, charts, and diagrams |
| 📄 Long context | Varies by tier and model — check official docs for current figures |
| 🤖 Tool use | Can call functions and external APIs in agentic workflows |
| 💻 Code | Strong across Python, TypeScript, JavaScript, Go, Java, C#, Rust |
| 🌍 Multilingual | Understands and generates in major world languages |
| 🛡️ Safety | Built-in Constitutional AI — designed to be helpful, harmless, and honest |
🎯 Which Claude Model to Use
Section titled “🎯 Which Claude Model to Use”| 🎯 Task | ✅ Best Model | 💡 Why |
|---|---|---|
| 🧩 The hardest reasoning/agentic tasks | Fable 5 | Anthropic’s most capable widely-released model |
| 🏗️ Architecture design or migration planning | Opus 5 | Deep reasoning + long context |
| 🚨 Production incident debugging | Opus 5 | Maximum reasoning on high-stakes problems |
| 📁 Large cross-file refactor | Opus 5 | Needs coherent multi-step planning |
| 👀 Risky pull request review | Opus 5 | Thorough bug detection across all files |
| 💻 Everyday coding and feature work | Sonnet 5 | Strong quality at reasonable cost |
| 📝 Writing, summarisation, documentation | Sonnet 5 | Excellent balance of quality and speed |
| 🤖 Production AI assistant (chat) | Sonnet 5 | Best default for most apps |
| 🏷️ Classification or extraction at scale | Haiku 4.5 | Fastest and cheapest for structured tasks |
| 🔀 Request routing | Haiku 4.5 | Simple judgment at low cost per call |
🌐 Where to Access Claude Models
Section titled “🌐 Where to Access Claude Models”| 🏷️ Platform | 🤖 Models Available |
|---|---|
| 🟣 Anthropic API (api.anthropic.com) | All current Claude models |
| 💬 Claude.ai | Opus, Sonnet, Haiku (product tiers) |
| 🟠 Amazon Bedrock | Claude Opus, Sonnet, Haiku |
| 🔵 Google Cloud Vertex AI | Claude Opus, Sonnet, Haiku |
| 🔀 OpenRouter | Claude models via unified API |