⚡ Your AI bill is eating your Q4 margin.Protect Your Margin. Get a FREE 360° Cost ReportFree 360° Cost Report

Home › Docs

Get started

apicuts Documentation

apicuts is an LLM API gateway with native OpenAI and Anthropic endpoints, prepaid workspace wallets, and per-token cost visibility. These docs cover setup, protocol compatibility, caching, and team spend controls.

How it works

  1. Sign up and generate a key. Create an account at /signup, fund your prepaid workspace wallet in /app/funding, and generate a company key (ack_...) in /app/keys — or explore the live demo or book a free 360° cost report first.
  2. Pick your base URL and model. Point OpenAI SDKs (Responses, Chat Completions, and OpenAI-compatible Gemini routes) at https://apicuts.ai/v1, or Anthropic SDKs and Claude Code at https://apicuts.ai. Fetch live workspace model IDs with GET /v1/models.
  3. Change one line (or run the companion). Update your SDK base URL, or install the macOS Companion V1 (curl -fsSL https://apicuts.ai/install | sh -s -- --key $APICUTS_API_KEY). See the quickstart.
  4. Track spend and token categories. Monitor input, cache-read, cache-write, and output tokens across your keys in your savings dashboard.

Why APICuts?

Native OpenAI & Anthropic endpoints

Change one base URL for OpenAI Responses (/v1/responses), Chat Completions (/v1/chat/completions), and native Anthropic Messages (/v1/messages). See the quickstart.

Prepaid workspace spend caps

Fund a prepaid workspace wallet and issue scoped ack_... keys per squad or repo so runaway loops stop cleanly instead of creating surprise month-end invoices. See spend controls.

Prompt-cache & coding-agent visibility

Forward Anthropic 5m and 1h prompt-cache breakpoints natively and audit cache-read vs cache-write ratios across coding tools. See the Claude Code team cost playbook.

Secure, transparent API access

Authorized providers only, prompts and outputs not stored by default, hashed company keys, and metadata-only request logging. See security and privacy.

OpenAI, Claude, Gemini and more

Official and authorized routes across OpenAI, Anthropic, and Google Gemini (via OpenAI-compatible /v1; native Gemini format is on the roadmap). See where traffic goes.

Real-time usage and baseline analytics

Track actual spend, token categories, and latency per request. See the savings dashboard and methodology.

Start here

Quickstart

Switch to apicuts with https://apicuts.ai/v1 or https://apicuts.ai, and roll back in one line too.

SDKs and integrations

OpenAI Responses & Chat Completions, Anthropic SDK, TypeScript, LangChain, LlamaIndex, Vercel AI SDK and cURL.

Coding tools

Claude Code, Codex CLI, Cursor, Continue, Cline, Aider and the macOS Companion V1.

Claude Code team costs

Repo guardrails, lean CLAUDE.md prefixes, /clear vs /compact, 5m cache TTLs, and prepaid wallet caps.

Routing

How requests are routed across configured models and providers.

Optimization techniques

Every lever we use: prompt/context caching, batching, prompt work, output limits and model routing.

Caching

Provider prompt, prefix and context caching with cache-aware routing.

Errors and retries

What each status code and error code means, idempotency rules, and which errors to retry.

Budgets, limits and alerts

Prepaid workspace wallets, concurrency caps, key scoping and spend alerts.

Savings dashboard

Actual spend, baseline costs and token-level usage visibility.

Reliability and status

Health endpoints, error handling, model catalog in /app/models (preview in demo) and SLAs.

Security and privacy

What we keep, what we never keep, and how requests and keys are handled.

Choosing an LLM gateway

How apicuts fits next to marketplaces, proxies and routers.

Tools and deep dives

Want a walkthrough of your team's API usage?

Get a free 360° cost report. Bring last month's provider usage and we'll show you your biggest waste and the first fix we'd ship.

Get a Free 360° Cost Report