Claude

Claude is a conversational AI assistant from Anthropic, used through the web app, mobile apps, and the API. It handles long-form writing, analysis, coding, document reading, and multi-step reasoning, with an emphasis on careful, steerable responses.

The current model lineup is Claude Fable 5.1 (the most capable widely released model), Claude Opus 5.5 for complex agentic coding and enterprise work, Claude Sonnet 5.5, and Claude Haiku 5.5, released October 7, 2026 as the first Haiku model with adjustable effort and now the default Haiku model in Claude Code. A separate Claude Mythos 5.1 exists but has limited availability through the invitation-only Cyber Verification Program.

All four current models, Fable 5.1, Opus 5.5, Sonnet 5.5, and Haiku 5.5, support a 1M-token context window, which helps with large codebases, long documents, or full transcripts in one conversation. The models also support multimodal input and tool use across the product.

Pricing is tiered. The Free plan is $0 and includes access to Sonnet and Haiku models, with usage caps. Pro is $20/month billed monthly, or $17/month if billed annually, and covers Opus, Sonnet, and Haiku plus Claude Code, with Fable 5.1 available through usage credits. Max starts from $100/month across 5x and 20x usage tiers, includes monthly API credits as of October 7, 2026, and allows Fable usage up to 50% of weekly limits; pricing above the $100 entry tier is not published. Team offers a Standard seat at $25/month billed monthly ($20/month billed annually) and a Premium seat at $125/month billed monthly ($100/month billed annually), with a two-seat minimum. Enterprise is $20/seat/month billed annually plus usage billed at API rates. API list pricing per million tokens is $10/$50 for Fable 5.1, $4/$20 for Opus 5.5, $2/$10 for Sonnet 5.5, and $0.10/$0.50 for Haiku 5.5 (prompts over 100K tokens: $0.50/$2.50).

Limitations are worth noting. Free and paid plans carry usage caps that reset on a rolling basis, so heavy users can hit limits mid-session. Max pricing beyond the $100/month entry tier isn't publicly disclosed, and the top-end Mythos 5.1 is invitation-only and not generally available. As with any LLM, outputs still need checking for factual accuracy.

Claude suits writers, developers, analysts, and teams who want strong reasoning, large-context handling across its current model lineup, and a choice of models to balance cost against capability.