Ethertext: Your AI Clipboard

One-Click Wonders

One click and your text transforms from good to great.

Customize Your Way

Choose the tone and style that fits your needs. Make your text yours.

Code Like a Pro

Explain, debug, or translate code snippets with laser precision.

Memorize and Recall

Let Ethertext memorize your text for later recall.

1. ⌘C: Copy

Copy your source text to your clipboard with ⌘C.

2. Click: Transform

Pick a transform from the dropdown from any of the Ethertext options.

3. ⌘V: Paste

After the beep, paste your text with ⌘V.

⌘⇧T: Clean text

Or, fix your selected text in one-shot with ⌘-shift-T.

⌘⇧M: Memorize

Memorize text or webpages with ⌘-shift-M.

⌘⇧D: Dictate

Record, transcribe and memorize your voice with ⌘-shift-D.

⌘⇧S: Capture screen

Take a screenshot and memorize the text on the screen with ⌘-shift-S.

⌘⇧R: Recall

Recall your past text with ⌘-shift-R and ask.

Add your OpenAI, Google Gemini, or Anthropic key

In Settings, add your keys from:

• OpenAI at https://platform.openai.com/account/api-keys
• Google Gemini at https://makersuite.google.com/app/apikey
• Anthropic at https://console.anthropic.com/settings/keys

Ollama support for local AI

Install Ollama and download some AI models, then enable Ollama support in Settings and pick your favorite. It's super quick with brew install ollama; ollama pull llama2 to get started if you're using Homebrew.

1.10 Quick and Mini (7/18/24)

Added OpenAI model: GPT 4o mini, which is the new default. Bug fixes and performance improvements.

1.9 The Next Top Models (6/22/24)

Added Anthropic Claude models: Opus, Sonnet, Claude. Added OpenAI model: GPT 4o. Added and updated Google Gemini models: 1.5 Pro, 1.5 Flash, 1.0 Pro. Added speaking style: "Talk like Gen Z."

1.8 Look at Me Now! (2/19/24)

Added Vision: take a screenshot and Ethertext will transcribe and remember, using Apple's local Vision model. Added Apple Audio model: using Apple's local Audio model is fast and free.

1.7 Talk to Me! (2/18/24)

Added Dictation: record your voice and Ethertext will transcribe and remember. Added embedding models: recall is vastly improved using OpenAI or Ollama for embeddings. Updated to latest OpenAI model versions (gpt-4-32k, gpt-4-0125-preview, gpt-3.5-turbo-0125). Prompts: "Improve text" grammar transforms optimized.

1.6 Creativity & Recency (1/30/24)

Added creativity (LLM temperature) setting. Added recent Q&A list. Added shortcut menu hints.

1.5 Shortcuts & Local Ollama (1/28/24)

Added Keyboard Shortcuts: you can now use commmand-shift-t to clean selected text in any app, or for quick-open memorize and recall. Added Ollama support: use a local model like llama2, Falcon, or Mixtral to transform text. Added latest GPT4 turbo model (gpt-4-turbo-preview) from Jan 25. New prompts: make efficient code, translate code to Javascript, and more/less formal text.

1.4 Memory (1/4/24)

Added Memory: you can now have Ethertext memorize and recall text for you. Added support for Google Gemini. Updated app icon to be rounded. New prompts: ELI4, add hype, and tone down.

1.3 Q&A (11/18/23)

Added Q&A: you can now edit your input to the prompts. Added Start at Login. Combined settings page includes version-checking about page.

1.2 Model Switching (11/6/23)

Added support for switching models to GPT4 Turbo Preview (released today), GPT4, GPT3.5 Turbo (new default), or DaVinci 003.

1.1 First Feedback (11/5/23)

Added custom prompt to settings. Made updating the API key not require a restart.

1.0 Initial release (10/29/23)