App Logo
claude.ai

List Claude.ai models and effort levels

List the models the Claude.ai composer offers, the effort levels available for the selected model, and which of each is currently selected. Models come back with their id, their display name and whether they sit in the picker's main list or under "More models". Effort levels are empty when the selected model has none (Haiku, for example, only offers an extended-thinking switch). Requires a signed-in session. Feed a model or effort id to set_model.

Input

None

Output

model required

Id of the selected model, e.g. "claude-opus-5".

effort required

Id of the selected effort level (low, medium, high, xhigh, max). Null when the selected model has no effort levels.

models required
effortLevels required

Effort levels the selected model offers, in the picker's order. Empty when it has none.

FAQ

List the models the Claude.ai composer offers, the effort levels available for the selected model, and which of each is currently selected. Models come back with their id, their display name and whether they sit in the picker's main list or under "More models". Effort levels are empty when the selected model has none (Haiku, for example, only offers an extended-thinking switch). Requires a signed-in session. Feed a model or effort id to set_model.
Nothing. It takes no input.
It returns model, effort, models, effortLevels.
Yes. It acts as you on claude.ai: on your own Chrome it reuses your session, and on a Reduck-hosted browser it loads the claude.ai cookies saved by the Reduck extension.
It only reads. It looks things up on claude.ai and changes nothing there.
Ask an AI agent connected to Reduck to run reduck/claude.ai/list_models, or run it from a terminal with the Reduck CLI: npx @reduck-ai/cli@latest run --script reduck/claude.ai/list_models
It is part of Reduck's official curated catalogue.

Related scripts

claude.ai

Ask Claude

Open a new Claude.ai chat, send one question, and wait for the answer to finish (up to 3 minutes). Returns the answer as the markdown the model emitted, the message and conversation ids, why generation stopped, which model answered, the web_search queries it issued, every source its web_search and web_fetch tools surfaced, and — separately — the sources the answer actually cited. Requires a signed-in session: claude.ai has no anonymous chat. Avoid running many of these at once: claude.ai limits how many chats one account may hold concurrently and refuses the extras, so a refused send comes back quickly carrying Claude's own message and only needs to be retried at a slower pace.

claude.ai

List Claude invoices

List every invoice across the orgs you have billing access to. Returns invoices (org_uuid, org_name, created_ts, due_date_ts, total, total_excluding_tax, currency, status, num_seats, hosted_invoice_url, invoice_pdf_url, payment_status) and skipped, newest first. Orgs you lack billing access to are reported in skipped rather than raising an error. Feed hosted_invoice_url to invoice.stripe.com/download_invoice_pdf to fetch the PDF (verified public/login-free, no Claude session needed).

claude.ai

Claude (Anthropic) login

Log into Claude (Anthropic) via Google SSO. Drives Claude's Continue-with-Google popup, picks the account (optional email arg), clears the consent screen, and lands in the Claude app. Runs on your local extension-paired browser only — it relies on the browser's own signed-in Google session, so it won't work on a cloud/managed browser without one.

Start building not clicking

Get started for free