# Acacus API

> Acacus is an AI assistant in Arabic and English, plus an OpenAI-compatible API at https://acacus.ly/v1, paid for in Libyan dinars. The API follows OpenAI's Chat Completions format, so the OpenAI libraries work with it once you change the base URL. You pay per token in Libyan dinars (LYD) from your prepaid API balance, with no foreign card, and a small free daily allowance lets you try it before you top up.

- Base URL: `https://acacus.ly/v1`. The earlier address, `https://ai.shafra.ly/v1`, is the same API and keeps working.
- Authentication: send your API key in the `Authorization` header: `Authorization: Bearer sk-shfr-...`.
- Models: `deepseek-v4-flash` (default), `deepseek-v4-pro`. `GET /v1/models` lists what your key can use.

## Get started

- [Overview](https://acacus.ly/docs): What the Acacus API gives you, its base URL, the two models it serves and how billing in Libyan dinars works.
- [Quickstart](https://acacus.ly/docs/quickstart): Create an Acacus key and send your first request with curl, Python or Node.js.
- [Authentication and API keys](https://acacus.ly/docs/authentication): Send your key in the Authorization header, and create, store and revoke keys.
- [Libraries and SDKs](https://acacus.ly/docs/libraries): Use the official OpenAI libraries, or plain HTTP, with the Acacus base URL.

## Guides

- [Streaming](https://acacus.ly/docs/streaming): Get the reply while it is being written, as Server-Sent Events.
- [Tool calling](https://acacus.ly/docs/tool-calling): Let the model ask your code to run a function, then answer with the result.
- [Images](https://acacus.ly/docs/images): Send images to deepseek-v4-flash in a user message.
- [Reasoning](https://acacus.ly/docs/reasoning): Turn DeepSeek's thinking step on or off, read the reasoning and see what it costs.
- [JSON output](https://acacus.ly/docs/json-output): Get replies that are valid JSON objects.
- [Prompt caching and cost](https://acacus.ly/docs/prompt-caching): How a repeated start of a prompt is read from the cache at a lower price.
- [Billing and the free tier](https://acacus.ly/docs/billing): Prices in LYD from your API balance, how each request is charged, and what the free daily allowance covers.

## API reference

- [Models](https://acacus.ly/docs/models): The two models the Acacus API serves, both running on DeepSeek V4 models (V4.1 Flash and V4 Pro), and GET /v1/models.
- [Chat completions](https://acacus.ly/docs/chat-completions): Every field of POST /v1/chat/completions and of its reply.
- [Errors](https://acacus.ly/docs/errors): The error format, every error code, and which errors to retry.
- [Rate limits and other limits](https://acacus.ly/docs/rate-limits): Requests per minute, requests at a time, request size and time limits.

## Help

- [Troubleshooting and FAQ](https://acacus.ly/docs/troubleshooting): Common problems and how to fix them.

## Optional

- [Changelog](https://acacus.ly/docs/changelog): Changes to the API, newest first.
- [All the docs in one file](https://acacus.ly/llms-full.txt): Every page above as Markdown, in reading order, with every code sample.