Auto Translator

Point MFK at any OpenAI-compatible /v1/chat/completions endpoint. Works with gemma-translate, Ollama, llama-server, OpenAI, Azure OpenAI, etc.

Provider	OpenAI-compatible chat/completions (Ollama, llama-server, OpenAI, Azure) gemma-translate (self-hosted TranslateGemma — POST /translate) DeepL (api-free.deepl.com / api.deepl.com) OpenAI-compatible: endpoint + model required. gemma-translate: endpoint required (e.g. https://ai-translate.60.internal.kr), model ignored. DeepL: API key only.
Endpoint URL
Model
API Key
DeepL Pro	Use api.deepl.com (paid). Leave unchecked for api-free.deepl.com.
Auto-translate	Automatically translate title / excerpt / content when a new translation is created Also publish the translated post (otherwise left as draft for review)
Translation memory	Cache translations — identical source strings skip the LLM round-trip

Save

Translation Memory

Cached entries	3
Cache hits (total)	0
Entries by target	en: 2  ja: 1

Clear translation memory

Usage

Once configured, use the CLI to auto-translate content:

wp mfk auto-translate post <post_id> --lang=en
wp mfk auto-translate term <term_id> --taxonomy=category --lang=en
wp mfk auto-translate strings --lang=en