Scholarly now supports Claude

Anthropic's Claude is now available in Scholarly's AI Chat. Two models join the lineup: Claude Sonnet 4.6 as the flagship reasoning model, and Claude Haiku 4.5 as a fast, cheaper everyday option.

Anthropic is the team behind Claude, and Claude has built a reputation for careful long-document reading, nuanced writing, and multi-step reasoning. If you've found yourself wanting a second opinion on a tricky proof, a more thoughtful editor for your essay, or a model that doesn't lose the thread halfway through a 200-page PDF, Claude is a real alternative to GPT and Gemini for studying.

Two Models, When to Pick Which

• Claude Sonnet 4.6 — the flagship. Strongest reasoning, deepest analysis of long PDFs and notes, currently in beta. Reach for it when you need the most thoughtful answer — dense readings, hard problem sets, careful essay feedback.
• Claude Haiku 4.5 — fast everyday chat. Cheaper, snappier, and still smart enough for most study questions. Great default when you're working through flashcards, summarising a lecture, or asking quick follow-ups.

Thinking Mode

Turn on Thinking in the chat menu and Claude will stream its reasoning before the final answer. It's especially handy on harder questions — you can watch the model work through a problem step by step, catch where its logic goes off, and learn from how it gets to the answer. It's like seeing the working, not just the result.

Works With the Rest of Scholarly

Claude isn't a sidecar — it plugs into everything else you already use. You can chat with Claude about your linked PDFs, recordings, podcasts, and pages. Web search, code execution, and file creation (PDFs, slide decks, and so on) all work the same way. Pick the model that fits the question; the rest of the workflow doesn't change.

Getting Started

Claude Sonnet 4.6 and Claude Haiku 4.5 are both on paid plans. To switch, open the model menu above the chat input and pick Claude Sonnet 4.6 or Claude Haiku 4.5. That's it — no settings to flip, no migration.

---

Give it a try on something hard — a tough chapter, a long research paper, a question you've been stuck on. Claude is genuinely good at the slow, careful kind of thinking that studying actually rewards. We'd love to hear what you make of it: [email protected].