arpk v0.3.0-beta.4

ARPK

LLM as your translator, with DeepLX-compatible API.

Usage

Prepare

ARPK supports most OpenAI-compatible APIs, Ollama is used here as an example (and is also the default):

ollama serve
ollama pull llama3.2

Install

First install Node.js, then install ARPK via your favorite package manager (or npm).

npm i -g arpk

You can also use ARPK via npx without installing it:

npx -y arpk

Translate

arpk translate --from "EN-US" --to "ZH" --input "Hello, World!"
# 你好，世界！

echo "Hello, World!" | arpk translate --from "EN-US" --to "JA"
# こんにちは、世界！

Serve

ARPK_LLM_MODEL="llama3.2" arpk serve

npx

ARPK_LLM_MODEL="llama3.2" npx -y arpk serve

docker

docker run -d \
  --name arpk \
  --network host \
  --restart unless-stopped \
  -e ARPK_LLM_MODEL=llama3.2 \
  ghcr.io/moeru-ai/arpk:latest

docker-compose

services:
  arpk:
    image: ghcr.io/moeru-ai/arpk:latest
    container_name: arpk
    network_mode: host
    restart: unless-stopped
    environment:
      - ARPK_LLM_MODEL=llama3.2

API

/translate, /api/v1/translate and /api/v2/translate are connected to the same translate endpoint.

await fetch('http://127.0.0.1:1188/translate', {
  body: JSON.stringify({
    source_lang: 'JA',
    target_lang: 'ZH',
    text: '雨の季節が過ぎ 澄み渡る空を 眺めて独り想フ'
  }),
  method: 'POST'
}).then(res => res.json())
// {
//   alternates: [],
//   code: 200,
//   data: '雨季过后，晴朗的天空下我独自遐思。',
//   id: 1519129853500,
//   method: 'ARPK v0.0.0 (llama3.2)',
//   source_lang: 'JA',
//   target_lang: 'ZH'
// }

Environments

Currently only Bearer Auth is supported when using ARPK_TOKEN, not URL Params.

License

MIT