Skip to content

Settings.MD

Latest commit

 

History

History
217 lines (160 loc) · 14.9 KB

Settings.MD

File metadata and controls

217 lines (160 loc) · 14.9 KB
Raw

Settings

These settings can be configured as environment variables to persist configurations across reinstallations.

Docker secrets

Any of the environment variables documented below can also be supplied by appending _FILE to the variable name and pointing it at a file containing the value.
When both <VAR> and <VAR>_FILE are set, <VAR>_FILE will be selected primarily.

Example:

services:
  lingarr:
    image: lingarr/lingarr:latest
    environment:
      OPENAI_API_KEY_FILE: /run/secrets/openai_api_key
    secrets:
      - openai_api_key

secrets:
  openai_api_key:
    file: ./secrets/openai_api_key.txt

General Configuration

Environment Variable Description
BASE_PATH Optional URL prefix to host Lingarr under a sub-path (e.g., behind a reverse proxy). Example: /lingarr. Leave unset to serve from the root.
MAX_CONCURRENT_JOBS Maximum number of Hangfire worker threads processing jobs in parallel. Defaults to 1 when unset.
JOB_TIMEOUT_MINUTES Minutes of silence before Hangfire retries a job. sqlite only, defaults to 30.
SQLITE_DB_PATH Path to the SQLite database file used when DB_CONNECTION is sqlite. Defaults to local.db within /app/config/.
DB_HANGFIRE_SQLITE_PATH Path to the SQLite database file used by the Hangfire job store. Defaults to /app/config/Hangfire.db.

Radarr Configuration

Environment Variable Description
RADARR_URL The base URL for the Radarr API. Example: http://localhost:7878.
RADARR_API_KEY The API key for authenticating with Radarr. Obtain this from your Radarr instance.

Sonarr Configuration

Environment Variable Description
SONARR_URL The base URL for the Sonarr API. Example: http://localhost:8989.
SONARR_API_KEY The API key for authenticating with Sonarr. Obtain this from your Sonarr instance.

Authentication Configuration

Environment Variable Description
AUTH_ENABLED Enables or disables authentication.
ENCRYPTION_KEYS The path for storing encryption keys, stored by default in the config directory.

Telemetry Configuration

Environment Variable Description
TELEMETRY_ENABLED Opt-in for anonymized telemetry to help other Lingarr users, defaults to false.

When enabled, Lingarr submits a report at most once every 7 days, and never from a development build. Each report contains the running Lingarr version, the platform, the report date, and the same lifetime totals shown on the Statistics page: files, lines and characters translated, plus the breakdowns by translation service, language pair, media type and model. No file names, subtitle content or other personal information is collected. Resetting statistics resets these totals as well.

Each report also carries an installation identifier. This is a random value generated once on first submission and stored locally as a setting; it is not derived from the machine, the network, the IP address or any hardware. The service that receives the report stores only a hash of it, never the original value, and uses it solely to tell one installation's reports apart from another's over time.

Translation Service Configuration

General Translation Settings

Environment Variable Description
SERVICE_TYPE An ordered JSON array of translation services. The first entry is the primary; the rest are fallbacks tried in order when the primary fails. A single value (e.g. openai) is still accepted and is normalised to a one-element array on startup.
SOURCE_LANGUAGES A minified JSON array of source languages for translation.
TARGET_LANGUAGES A minified JSON array of target languages for translation.
USE_BATCH_TRANSLATION Set to true to translate subtitles in batches with AI services, defaults to false.

The SOURCE_LANGUAGES and TARGET_LANGUAGES variables should be provided as a minified JSON array. Each object in the array should contain the name and code of the language.

Example:

[{"name":"English","code":"en"},{"name":"Dutch","code":"nl"}]

Service type

LibreTranslate

Environment Variable Description
LIBRE_TRANSLATE_URL The base URL for the LibreTranslate API. Example: https://libretranslate.com.
LIBRE_TRANSLATE_API_KEY The API key to use for the LibreTranslate API. This is optional, and only needed if using an instance that requires one.

OpenAI

Environment Variable Description
OPENAI_MODEL The model to use for OpenAI translations. Example: gpt-4, gpt-3.5-turbo.
OPENAI_API_KEY The API key for authenticating with OpenAI.
AI_PROMPT The prompt template for AI-based translation services.
AI_USER_PROMPT The user message template. Supports {lineToTranslate}, {contextBefore} and {contextAfter}.

Anthropic

Environment Variable Description
ANTHROPIC_MODEL The model to use for Anthropic translations. Example: claude-opus-5.
ANTHROPIC_API_KEY The API key for authenticating with Anthropic.
ANTHROPIC_VERSION The version of the Anthropic API to use.
AI_PROMPT The prompt template for AI-based translation services.
AI_USER_PROMPT The user message template. Supports {lineToTranslate}, {contextBefore} and {contextAfter}.

Gemini

Environment Variable Description
GEMINI_MODEL The model to use for Gemini translations. Example: gemini-2.0-flash.
GEMINI_API_KEY The API key for authenticating with Gemini.
AI_PROMPT The prompt template for AI-based translation services.
AI_USER_PROMPT The user message template. Supports {lineToTranslate}, {contextBefore} and {contextAfter}.

DeepSeek

Environment Variable Description
DEEPSEEK_MODEL The model to use for DeepSeek translations. Example: deepseek-chat.
DEEPSEEK_API_KEY The API key for authenticating with DeepSeek.
AI_PROMPT The prompt template for AI-based translation services.
AI_USER_PROMPT The user message template. Supports {lineToTranslate}, {contextBefore} and {contextAfter}.

Mistral

Environment Variable Description
MISTRAL_MODEL The model to use for Mistral translations. Example: mistral-large-latest.
MISTRAL_API_KEY The API key for authenticating with Mistral.
AI_PROMPT The prompt template for AI-based translation services.
AI_USER_PROMPT The user message template. Supports {lineToTranslate}, {contextBefore} and {contextAfter}.

xAI

Environment Variable Description
XAI_MODEL The model to use for xAI translations. Example: grok-4.5.
XAI_API_KEY The API key for authenticating with xAI.
AI_PROMPT The prompt template for AI-based translation services.
AI_USER_PROMPT The user message template. Supports {lineToTranslate}, {contextBefore} and {contextAfter}.

LocalAI

Environment Variable Description
LOCAL_AI_MODEL The model to use for LocalAI translations.
LOCAL_AI_API_KEY The API key for authenticating with LocalAI.
LOCAL_AI_ENDPOINT The full URL of the completion endpoint. Example: http://ollama:11434/v1/chat/completions.
LOCAL_AI_STRUCTURED_OUTPUT Send a JSON schema with batch requests to a chat/completions endpoint. Default: true. Set to false for endpoints that accept the schema but do not support it, such as DeepSeek models.
AI_PROMPT The prompt template for AI-based translation services.
AI_USER_PROMPT The user message template. Supports {lineToTranslate}, {contextBefore} and {contextAfter}.

Proofreading

Proofreading is only available for the AI services above (OpenAI, Anthropic, Gemini, DeepSeek, Mistral, xAI, LocalAI), and only for a service_type entry that supports it. It uses two more settings, shared across all AI services the same way AI_PROMPT and AI_USER_PROMPT are. These are configured through Settings > Services > Request Settings in the web interface, or through their environment variables:

Setting Environment Variable Default
proofread_prompt PROOFREAD_PROMPT You are proofreading a subtitle translated from {sourceLanguage} to {targetLanguage}. Compare the translation against the source and correct mistranslations, wrong names, grammar and punctuation, preserving the tone and meaning without censoring the content. Keep the length close to the original so it still fits on screen. If the translation is already correct, repeat it unchanged. Provide only the corrected translation as output, with no additional comments.
proofread_user_prompt PROOFREAD_USER_PROMPT [SOURCE] {sourceLine}\n[TRANSLATION] {translatedLine}

See the AI services documentation for the full placeholder list.

DeepL

Environment Variable Description
DEEPL_API_KEY The API key for authenticating with DeepL.

Translation Services

The SERVICE_TYPE variable determines which translation services to use. Provide a JSON array, the first entry is the primary service and any remaining entries are fallbacks, tried in order when the primary fails. A single value is also accepted and is rewritten to an array on startup.

Example:

["openai","anthropic","libretranslate"]

The supported values are:

Value Description
libretranslate Use the LibreTranslate service for translations.
openai Use OpenAI's models (e.g., GPT-4, GPT-3.5) for translations.
anthropic Use Anthropic's models (e.g., Claude-2) for translations.
gemini Use Google's Gemini models for translations.
deepseek Use DeepSeek's models for translations.
mistral Use Mistral's models for translations.
xai Use xAI's Grok models for translations.
localai Use a locally hosted or OpenAI-compatible AI model for translations.
deepl Use the DeepL API for translations.
google Use Google Translate for translations.
bing Use Bing Translate for translations.
microsoft Use Microsoft Translator for translations.
yandex Use Yandex Translate for translations.