ew-translator

Instant, lightweight popup translator right at your fingertips.
Select text in any application, press a global shortcut, and view the translation instantly next to your pointer.

Why ew-translator?

• ⚡ Lightning Fast & Pure Rust — Instant startup and minimal memory footprint. No heavy web engines, Electron, or WebKit.
• 🎯 Smart Adaptive Popup — The window automatically hugs the content. A tiny badge for single words, a comfortable card for sentences, and a smooth scrollbar for long articles.
• 🖥️ Multi-Monitor & Edge Aware — Never goes off-screen. Intelligently flips upward when invoked near the bottom of your display and stays clamped within your monitor's bounds.
• ✨ Seamless UX — Copy translated text with standard selection (Ctrl+C). Dismiss with a single click outside or Esc.
• 🤖 Modern AI & Traditional Providers — Works out-of-the-box with Google Translate (zero setup), or connect Google Gemini, OpenRouter, OpenAI, or local LLMs (Ollama, LM Studio).

---

Quick Start

Launch ew-translator with your preferred hotkey and target language:

# Default: Russian translation with F7 hotkey
ew-translator -h f7

# Translate into English with Alt+T
ew-translator -l en -h 'ALT+T'

Select text anywhere on your screen and press your hotkey!

---

Translation Providers

1. Google Translate (Default)

Ready to use immediately with zero configuration:

ew-translator -h f7

2. Google Gemini

Translate using Gemini's fast models (defaults to gemini-2.5-flash-lite):

export EW_TRANSLATOR_SOURCE='gemini'
export EW_TRANSLATOR_API_KEY='your-gemini-key'
ew-translator -h f7

3. OpenRouter

Access hundreds of models (e.g. google/gemini-2.5-flash-lite, Claude, Llama):

export EW_TRANSLATOR_SOURCE='openrouter'
export EW_TRANSLATOR_API_KEY='your-openrouter-key'
export EW_TRANSLATOR_MODEL='google/gemini-2.5-flash-lite'
ew-translator -h f7

4. OpenAI & Local LLMs

Use OpenAI directly or any OpenAI-compatible endpoint (like Ollama or LM Studio):

export EW_TRANSLATOR_SOURCE='openai'
export EW_TRANSLATOR_API_KEY='your-key'
export EW_TRANSLATOR_MODEL='gpt-4o-mini'
# For local LLMs, add your API URL:
# export EW_TRANSLATOR_API_URL='http://localhost:11434/v1'
ew-translator -h f7

---

Installation

Prebuilt Binaries

Download the latest prebuilt binary from Releases:

install -m 755 ew-translator ~/.local/bin/ew-translator

Build from Source

Ensure you have Rust and Cargo installed:

git clone https://github.com/bnku/ew-translator.git
cd ew-translator
cargo build --release
install -m 755 target/release/ew-translator ~/.local/bin/ew-translator

---

Configuration

Settings can be specified via CLI flags, environment variables, or an optional configuration file (~/.config/ew-translator/config.toml):

Option	Environment Variable	Description	Default
-l, --lang	EW_TRANSLATOR_LANG	Target language code	ru
-h, --hotkeys	EW_TRANSLATOR_HOTKEYS	Global activation shortcut	CTRL+SHIFT+F7
-s, --source	EW_TRANSLATOR_SOURCE	Provider (google-translate, gemini, openrouter, openai)	google-translate
-m, --model	EW_TRANSLATOR_MODEL	Model ID (for AI providers)	Provider default
-u, --api-url	EW_TRANSLATOR_API_URL	Base API URL	Provider default
	EW_TRANSLATOR_API_KEY	Provider API key	None
-c, --config	EW_TRANSLATOR_CONFIG	Path to custom config file	~/.config/ew-translator/config.toml

Optional Config File Example

~/.config/ew-translator/config.toml:

source = "openrouter"
api_key = "sk-or-v1-..."
model = "google/gemini-2.5-flash-lite"
lang = "ru"
hotkeys = "F7"

---

License

GPL-3.0