Claude Models: Full List, Key Features & Use Cases

Anthropic's Claude model family has grown from a single 2023 chatbot into a multi-tier lineup spanning Haiku, Sonnet, Opus, and now a new Mythos class positioned above Opus. This guide lists every major Claude model in order from the newest to the oldest, with a concise overview of each model's specs, key features, and intended use — plus links to try the ones available on Chat Smith.

Released June 9, 2026, Claude Fable 5 is Anthropic's first publicly available Mythos-class model, a new tier positioned above Opus. It shares its underlying model with Claude Mythos 5, but ships with safety classifiers that reroute high-risk cybersecurity, biology, and chemistry requests to Claude Opus 4.8. Mythos 5 has some of those safeguards lifted and stays restricted to vetted partners. Context window: 1M tokens. Pricing: $10/1M input, $50/1M output. Best for: the hardest, longest coding and knowledge-work tasks.

Released June 30, 2026, Claude Sonnet 5 replaced Sonnet 4.6 as Anthropic's default mid-tier model, delivering near-Opus performance on agentic coding, computer use, and tool use at a lower price. Best for: everyday coding, agents, and general-purpose work where Opus-level cost isn't justified.

Released May 27–28, 2026, Claude Opus 4.8 is Anthropic's Opus-tier flagship, roughly 4x less likely than Opus 4.7 to let flaws in its own code pass unremarked, and it scores 84% on Online-Mind2Web. Context window: 1M tokens by default. Pricing: $5/1M input, $25/1M output. Best for: long-running autonomous coding agents and complex multi-step engineering work.

Released April 2026, Claude Opus 4.7 built on Opus 4.6's coding and agentic strengths ahead of the 4.8 refresh two months later. Best for: reference only — fully superseded by Opus 4.8.

Claude Opus 4.6 launched February 5, 2026 with stronger code review, debugging, and planning, plus a 1M-token context window in beta. Claude Sonnet 4.6 followed on February 17 as the everyday-use sibling, scoring 79.6% on SWE-bench Verified — among the highest of any model at its release. Best for: agent planning, design, and coding at a lower cost than Opus (Sonnet 4.6); reference only for Opus 4.6, now superseded.

Released November 24, 2025, Claude Opus 4.5 focused on spreadsheets, long-running chats, and high-intelligence coding. Best for: reference only — superseded by the Opus 4.6/4.7/4.8 line.

Released October 15–16, 2025, Claude Haiku 4.5 brought coding, reasoning, and computer-use capability to Anthropic's fastest, most affordable tier. Best for: high-volume tasks, latency-sensitive applications, and cost-conscious agent workflows.

Released September 29, 2025, Claude Sonnet 4.5 focused on improving coding, computer use, and agentic workflows over Sonnet 4. Best for: reference only — superseded by Sonnet 4.6 and Sonnet 5.

Released August 2025, Claude Opus 4.1 was an incremental refinement of Opus 4, improving coding and reasoning ahead of the more significant 4.5 upgrade. Best for: reference only — fully superseded.

Released May 2025, Claude Opus 4 and Claude Sonnet 4 became the foundation for every later 4.x upgrade, pushing Claude deeper into coding, agentic work, and tool use. Both were retired on June 15, 2026. Best for: historical reference only.

Released February 2025, Claude 3.7 Sonnet was Anthropic's first Claude model with a visible extended-thinking mode, letting it reason step by step before answering. It was retired February 19, 2026. Best for: historical reference only.

Claude 3.5 Sonnet launched June 2024 and got a significant "v2" update in October 2024 alongside Claude 3.5 Haiku, adding computer-use capability for the first time. Most of the 3.5 generation has since been retired, though some providers still list Claude 3.5 Haiku. Best for: historical reference — verify current availability before building on these models.

Released March 2024, Claude 3 introduced the Opus/Sonnet/Haiku tier structure Anthropic still uses today, along with vision input and a big jump in reasoning quality over Claude 2. All three models have since been deprecated. Best for: historical reference only.

Claude 2 launched July 2023 with a much larger context window than Claude 1, and Claude 2.1 followed in November 2023 with improved accuracy and a 200K-token context window — the standard every Claude model since has kept. Best for: historical reference only.

Released March 2023, Claude 1 was Anthropic's first public model and the starting point for the entire Claude lineup. Best for: historical reference only.

Anthropic's Claude lineup now spans four capability tiers, from the fast, affordable Haiku line to the new Mythos class above Opus. For teams and individuals who want to use these models without managing a separate Anthropic account, Chat Smith gives you access to the available Claude models alongside GPT, Gemini, Grok, and DeepSeek — all from a single interface. Browse the full model catalog on Chat Smith to find the right model for your workflow.