Which Claude Model to Use: Fable, Opus, Sonnet or Haiku
What Claude Fable 5.2, Opus 5.5, Sonnet 5.5 and Haiku 5.5 are for, what each costs, which plans include them, and when Anthropic says to move up.

What Claude Fable 5.2, Opus 5.5, Sonnet 5.5 and Haiku 5.5 are for, what each costs, which plans include them, and when Anthropic says to move up.

Claude Fable 5.2, Gemini 4 Carbon and GPT-6 Astra compared on API price, context limits, app access and one shared benchmark table from Anthropic.

Anthropic added Sheets, Whiteboards and Data to Claude, plus apps that open inside chat. Here is what each does, who gets it, and where Meetings stands.

Anthropic set a $70 to $78 IPO range for 825M shares, with $26B more from Google, Nvidia and five others. Bloomberg says orders top the offer.

Anthropic says Claude Fable 5.2 designed a speakerphone, a voice hub and a virtual pet with Diode, and the boards worked with no circuit fixes.

Anthropic's Claude Fable 5.2 costs $8/$40 per million tokens, down from $10/$50, runs up to 30% faster, and tops Opus 5.5 on most of Anthropic's tests.

Claude Fable 5.2 and Mythos 5.2 launched with a 20% price cut. Testers, partners and Hacker News weigh in on long agent runs, cost per task and nerf fears.

Hankyung reports a planned 300MW AI data center in Gumi, South Korea, fell through after lease talks with Anthropic ended over site issues.

Claude Code Projects, in public beta on Pro, Max and Ultra, let Claude coordinate a stream of tasks as parallel cloud threads. How it works, usage and limits.

Startup PSI says a non-public Anthropic model found a Lean-checked proof of the Aaronson-Ambainis conjecture. Scott Aaronson believes it is correct.

Matteo Raso's essay doubting the AI bubble drew 756 Hacker News comments. Commenters push back on its data center, profit and moat claims; some agree.

A viral line says Amodei predicts half of entry-level lawyers will be wiped out. What he said, and what hiring data shows.

A UChicago law professor says Claude deleted his syllabus and a year of notes during a page-number fix. Dropbox saved them. What to know.

Claude Code 2.1.319 moves long MCP server jobs to the background in the terminal and adds a setting that denies tool calls instead of waiting on prompts.

Ten copy-paste Claude prompts for research and document analysis, built from Anthropic's docs, plus what happened when we tested them on a small local model.

Anthropic's agent memory research says newer Claude models judge better what to remember, and a dreaming pass over transcripts can fix wrong notes.

The FT reports Anthropic agreed to pay about $3.4 billion for 1.7 GW of OnePWR gas generators and the capability to install them at its data centers.

A viral X post said Claude Opus 5.5's reasoning changed on November 3. Theo's data points to a new thinking summarizer, and HN and X are still split.

Claude's Personal Finance beta links US bank and card accounts through Plaid, read-only. What it sees, how privacy works, and its current limits.

A judge let three claims proceed in Kahn v. Anthropic over the Claude Max 5x or 20x usage pitch and dismissed four with leave to amend.
