Why 8.5/10: Top pick for writing: prose that many writers describe as natural, very long context, Projects for organising long work and a capable free tier. Points off for usage limits that are hard to predict and for weekly caps on paid plans.
By ToolSift editorial. Last checked . Prices change often; "check site" means verify with the vendor.
What it does
Claude is Anthropic's AI assistant, available on the web, desktop and mobile. Anthropic's pricing page lists, even on the free plan: web search, creating files and running code, memory across conversations, connectors to apps and tools, artifacts, and Claude Design, Slides and Docs. Paid plans add more usage, more models (Opus on Pro and above), unlimited Projects (free is capped at 5), Research, Claude Code, Claude in Chrome and Claude for Microsoft 365.
For writers, the useful pieces are Projects (a workspace holding reference files and standing instructions for one piece of work), artifacts (a side panel for a document you're revising), and the long context window, which Anthropic lists as up to 1M tokens depending on the model. That lets you work across a long report, a book chapter or a stack of interview transcripts in one conversation.
Anthropic publishes how limits work: every plan resets on a rolling five-hour session window, paid plans add weekly limits, and chats and Claude Code share one pool. Pro gives at least 5x the free usage per session, and Max gives 5x or 20x Pro. Paid users can turn on usage credits to keep working at API rates. Model training on consumer plans is opt-out, per the plan table.
Against our runners-up: ChatGPT is the better all-rounder if you want image generation, voice and a huge plugin ecosystem in one place, and Grammarly is the lighter choice for polishing what you've already written inside other apps. Claude's edge for writers is the tone of its drafts and how comfortably it works through long source material, which makes it a good partner for reports, essays and long articles rather than quick one-line edits.
Our verdict
Claude, from Anthropic, is our pick for writing because its prose is often described as natural and less formulaic and handles long documents well, with a context window of up to 1M tokens depending on model. Projects keep the sources and instructions for a long piece in one place, and artifacts make drafting and revising comfortable. The free plan is genuinely useful; Pro at $20 a month adds more usage, Projects without the free cap, Research and Claude Code. The main frustration is limits: usage resets on rolling five-hour windows plus weekly caps, with no fixed message count.
✅ Pick it if You care about voice, tone and long-form drafts, and want to work across long source documents in one place.
⛔ Skip it if You need a fixed, predictable message allowance, or your workflow centres on built-in image generation.
Pricing: As of 19 Oct 2026 (USD, excluding tax): Free $0; Pro $20/month, or $17/month billed annually ($200 up front); Max from $100/month (5x Pro usage) or 20x usage (the page lists Max 5x and Max 20x tiers, with $100–$200/month in API credits included); Team for 2–150 people; Enterprise $20 per seat per month plus usage at API rates, billed annually. Check site.
Claude is built by Anthropic, PBC, a US AI company.
What to watch
Limits are rolling, not fixed: five-hour windows plus weekly caps on paid plans, and Anthropic may set other limits at its discretion. Long documents and heavier models use the allowance faster.
Training: model training on Free, Pro and Max is opt-out per Anthropic's plan table, so check your privacy settings if you paste client or confidential material. Team and Enterprise have no training by default.
Refunds are generally not offered except where terms or local law require; UK and EEA users have a 14-day withdrawal window, per the FAQ.
Facts still need checking: like any language model, Claude can state things confidently that are wrong. Verify names, numbers and quotes before publishing.
Free Projects are capped at 5; heavy long-form users will likely need Pro.