EnviousWispr

Talk naturally. Paste perfectly.
Free, on-device AI dictation and speech-to-text for macOS.
Powered by Apple Silicon. No cloud, no account, your voice never leaves your Mac.

Demo

https://github.com/user-attachments/assets/2655d632-1ad8-4a00-bac8-d6c3cc2f6aad

What is this?

EnviousWispr is a free AI dictation app for macOS that runs entirely on-device. It uses Whisper and Parakeet speech-to-text models on Apple Silicon to transcribe your voice locally, polishes the output with an optional LLM, and pastes clean text into whatever app you're working in. Transcription is sub-second; with optional AI polish, the full keybind-to-paste flow typically lands in around a second and a half.

No cloud. No account required. No subscription. No audio ever leaves your Mac. Works fully offline.

It is open source under the GPLv3, actively maintained, and built to be a tool you can rely on every day.

Why EnviousWispr?

| | EnviousWispr | Cloud dictation services | |---|---|---| | Privacy | 100% on-device transcription | Audio uploaded to servers | | Speed | Sub-second transcription, paste-on-stop | Network round-trip latency | | Models | Parakeet Ultra (NVIDIA NeMo) + OpenAI Whisper (via transcribe.cpp) | Single vendor model | | Polish | Optional. Fully on-device (EG-1, S1-mini by Superwhisper, Apple Intelligence, Ollama) or bring-your-own-key cloud (OpenAI, Gemini, Claude) | Cloud polish, included in subscription | | Cost | Free. No account, no subscription | Monthly subscription | | Works offline | Yes, fully functional without internet | No |

How it works

Press keybind -->  Record  -->  Transcribe  -->  Polish (optional)  -->  Paste
    ~0ms          live        ~400-800ms         ~200-500ms            instant

1. Press your keybind from any app. Push-to-talk, toggle, or hands-free (double-press to lock for long-form), your choice.
2. Speak naturally. Silero VAD detects when you stop talking and ends recording automatically.
3. On-device transcription. Choose Parakeet Ultra (fastest, 25 European languages) or Whisper (99+ languages, with automatic language detection).
4. AI polish (optional). Clean up grammar, punctuation, and formatting. Runs fully on-device with EG-1 (our own custom model), S1-mini by Superwhisper, Apple Intelligence (macOS 26+), or Ollama, or in the cloud via OpenAI, Gemini, or Claude with your own API key.
5. Text lands in your clipboard and optionally auto-pastes into the active app.

See the full interactive pipeline demo at enviouswispr.com/how-it-works

Reliability is a feature

Dictation is only useful if you can trust it mid-sentence, every time. EnviousWispr keeps the critical path (record, transcribe, paste) deliberately separate from every optional enhancement, so a hiccup in a "nice to have" can never swallow your words. When an optional step cannot run, you simply get your raw transcribed text instead of an error.

| What we hardened | What it means for you | |---|---| | Delivery survives non-critical failures | If saving to your history cannot complete (full disk, permissions), your dictation is still pasted. The save is best-effort and never blocks delivery. | | Paste that actually lands | A multi-step delivery path tries the fastest reliable method first and falls back automatically, so text lands even in apps that resist the usual paste (Word, Excel, Pages, Numbers, and more). | | Onboarding that won't leave you half-set-up | Setup won't let you start until Accessibility is granted, and it re-checks if you later revoke permission. | | Clear answers when AI polish has a problem | If a cloud or local model fails (OpenAI, Gemini, Ollama), you get a specific, plain-language message, and your raw text still arrives. | | Deterministic cleanup before AI | For English, numbers, dates, and money are formatted by a fixed, predictable step, even when AI polish is off or unavailable. | | No silent freeze after idle | The speech engine runs inside EnviousWispr itself, not in a separate helper macOS could shut down behind its back. A press after a long idle no longer stalls for a few seconds with nothing on screen; if the model was unloaded by your idle setting, the pill says it is getting dictation ready. | | Privacy-safe diagnostics | Crash reports carry counts and context, never your dictations, transcripts or audio, and are redacted before they are sent. | | Hardened releases | Every build is signed, notarized, and Gatekeeper-checked before it ships. |

Supported Models

| Model | Best for | Languages | Disk space | Runs on | |---|---|---|---|---| | Parakeet Ultra | Fastest dictation (default) | 25 European languages | ~630 MB | Apple Neural Engine | | Whisper (Whisper Large v3 Turbo) | Broadest language coverage and automatic language detection | 99+ languages | ~0.9 GB | Apple GPU |

Both models run entirely on-device on Apple Silicon. Parakeet runs on the Apple Neural Engine through Core ML, which is what makes the default engine near-instant; Whisper runs on the GPU through transcribe.cpp (Metal) for broad-language accuracy. Models are downloaded separately and stored on your Mac.

On-device AI polish

Transcription gets your words down. AI polish cleans them up: it drops filler, fixes grammar and punctuation, and structures rambling speech into readable text. This step is optional, and by default it never leaves your Mac.

| Polish engine | What it is | Runs on | Extra download | |---|---|---|---| | EG-1 (recommended) | Our own model, custom fine-tuned for dictation cleanup | On-device, macOS 14+ | ~2.9 GB (optional) | | S1-mini by Superwhisper | A small open model for dictation cleanup, happiest in English, with Tone, Structure, and Context writing style settings | On-device, macOS 14+ | ~484 MB (optional) | | Apple Intelligence | Apple's on-device model, no extra download | On-device, macOS 26+ | none | | Ollama | Use a model on your Mac or one hosted by Ollama | On-device or Ollama's servers | varies for local models; none for hosted | | OpenAI / Gemini / Claude | Bring-your-own-key cloud polish, text only | Cloud (your key) | none |

EG-1 is our own AI model, fine-tuned specifically for dictation cleanup and optimized for Apple Silicon. It runs entirely on your Mac with no internet required, and it closes the gaps a general on-device model leaves: reliably turning a spoken list into a real list, splitting a wall of speech into clean paragraphs, and keeping only the corrected version when you fix yourself mid-sentence. Because it is our own model rather than Apple's, it works across the full supported range (macOS 14 and later), not just macOS 26. EG-1 is distributed under its own model license, not the GPLv3 that covers the app code (see License).

On our own benchmark of 1,890 real dictation-cleanup cases, EG-1 passed 93.7%, ahead of both GPT-5.4-mini (83.8%) and Gemini 3.5 Flash (92.6%) on the same cases with the same judge. This is our own benchmark, not an independent review. The eval harness and the exact prompts are public in scripts/eval/ so you can inspect or rerun them; the test cases are personal dictations and stay private.

Features

• 🎙️ Dual ASR engines with Parakeet Ultra (NVIDIA NeMo) and OpenAI Whisper on transcribe.cpp
• 👀 Live Preview: watch your words appear in the recording pill while you speak, so you can see it is working without waiting for the paste. On by default, switchable off, and it never changes the text you get. A Mac that cannot run it behaves as though it were off
• ↩️ Escape Recovery: hit your cancel keybind by mistake and get the dictation back. On by default: a keybind cancel keeps the text for 24 hours instead of discarding it, and you can switch that off. The Cancel button in the recording pill still discards on the spot, so you keep one way to mean it. Only the text is kept, never the audio
• ✨ AI polish that respects your words: strips filler words and false starts, fixes grammar and punctuation, formats numbers, dates, and URLs, and honors your custom vocabulary, all in your spoken language (never translated or rewritten)
• 🔒 Polish that can stay private: run it fully on-device with EG-1 (our own custom model), S1-mini by Superwhisper, Apple Intelligence (macOS 26+), or Ollama, or in the cloud via OpenAI, Google Gemini, or Claude with your own API key
• ✂️ Snippets: say a keyword and a short phrase, and the text you saved is pasted word for word, with any requested date, time or clipboard fill-ins filled in, never polished. An email address, a sign-off, a link you send every week
• 📁 Transcribe a File: turn one audio or video file you already have (m4a, mp3, wav, aiff, caf, mp4, mov or flac) into clean text. When a recording has more than one voice and the app can separate the voices and match the words to them, the transcript comes back as speaker turns. The audio and the speaker detection stay on your Mac; cleanup follows the polisher you pick
• 🌍 Multilingual with automatic language detection: speak in any supported language and EnviousWispr detects it, then offers to lock it in for faster, more accurate transcription
• 😀 Speak an emoji: say the emoji's name followed by "emoji" (like "thumbs up emoji") and the glyph drops right in
• ✋ Voice Activity Detection via Silero VAD that stops recording automatically when you stop talking
• 📚 Custom vocabulary and vocabulary packs for names, brands, and technical terms the ASR might miss, plus one-tap import of names from your Contacts (which never leave your Mac)
• 🧠 Self-Learning Dictionary: correct a dictation and EnviousWispr detects the change and adds the intended term to your dictionary
• 📋 Paste Last Dictation: ⌃⌘V pastes your most recent dictation wherever your cursor is and ⌃⌘C copies it, both changeable in Keybinds
• 🔇 Media while you dictate: choose whether your music and videos continue, lower, mute or pause while you talk
• 🇩🇪 English and German interface: choose the app's language in App Settings > Appearance
• 💬 Send Feedback: tell us what works from inside the app, and it may suggest help articles that answer your question before you send
• 🛡️ Privacy controls: choose whether to share usage metrics and crash reports in App Settings > Privacy
• ➕ Quick Add: highlight a misheard word anywhere on macOS and save the right spelling to your dictionary with a keyboard shortcut or from the menu bar, without opening Settings
• 🎨 Your choice of recording pill: pick the recording indicator design you like in Dictation Settings > Recording Pill, and try a practice dictation before setup ends
• ⌨️ Global keybind with push-to-talk, toggle, and hands-free modes (double-press to lock for long-form dictation)
• 📋 Auto-paste directly into the active app, or just copy to clipboard
• 🕘 History for browsing, searching, and reviewing past dictations and transcripts, with an All / Dictations / Transcripts filter above the list
• 🔎 Settings search: type in the Settings sidebar, or press ⌘F, to jump straight to any setting
• 🧭 Menu bar native with minimal footprint
• 🔄 Auto-updates via Sparkle

Recent improvements

EnviousWispr ships often. A few of the user-facing improvements from recent releases:

• Fixed a Crash at Launch. Fixed a crash at launch for some people using EnviousWispr in German. (v2.5.5)
• Smoother Quitting. More reliable quitting after using Whisper or Universal Live Preview, and less memory in use. (v2.5.5)
• Pills Fit Long Messages. Long pill messages now wrap across more lines. (v2.5.5)
• Improved Dictation Engine: Parakeet Ultra. In our tests across 24 languages, Parakeet Ultra made about 19% fewer word errors than the old Parakeet, at the same speed. It downloads and switches over on its own. (v2.5.4)
• Improved Whisper Engine. Whisper now runs on a new engine built for Apple Silicon. In our tests it made about 13% fewer errors, transcribed long recordings more than twice as fast, and stayed steady through a full day of dictation. (v2.5.4)
• Introducing "Hold plus Space". In Push to Talk with a modifier key such as Right Option, hold it and press Space to go hands-free. (v2.5.4)
• Introducing "Casual mode". Leave off the full stop at the end of each dictation. (v2.5.4)
• Say More Symbols. Say brackets, quotes, ampersands and more. (v2.5.4)
• Introducing "Keep Out of Clipboard History". A new switch asks clipboard history apps to skip your dictations. (v2.5.4)
• Smoother Every Day. Steadier record key, Bottom pills above the Dock, better Chrome pasting and fewer crashes. (v2.5.4)
• Streamlined Settings with Instant Search. Completely redesigned the settings UI to eliminate clutter, featuring a new search bar to quickly jump to any configuration option. (v2.5.3)
• Modernized Menu Bar. Refreshed the menu bar interface with a cleaner layout and faster access to primary controls. (v2.5.3)
• Launcher Support (Raycast). Dictation now works inside Raycast. (v2.5.3)
• Quiet Update Notifications. Added a toggle to prevent the menu bar icon from changing states when an update is available. (v2.5.3)
• Spoken Punctuation (DE, FR, ES, IT). Added spoken punctuation support for German, French, Spanish, and Italian using localized trigger words (e.g., "Diktiere Punkt" in German). (v2.5.3)
• Expanded Regional Formatting. Dictated phone numbers, timestamps, and email addresses now format accurately in more European languages. (v2.5.3)
• Faster Local Polishing with EG-1. Reduced latency when cleaning up and formatting longer dictations using the on-device EG-1 engine. (v2.5.3)
• Updated Model Support. Added support for Claude 5.5 and OpenAI gpt-6, alongside significant speed improvements when using Gemini 3.8 Flash. (v2.5.3)
• Smarter Self-Learning Dictionary. Tuned phrase detection to apply learned vocabulary fixes more consistently. (v2.5.3)
• Escape Recovery & Undo. Undo after Escape now pastes your words into the field you were in, or copies them to your clipboard if it can't paste. (v2.5.3)
• Privacy controls. Choose whether to share usage metrics and crash reports in Settings > Permissions. Crash report changes apply after a restart. When sending feedback, choose whether to attach diagnostics and preview the file first. (v2.5.2)
• Word check loads on demand. Envious Word Check now loads when a dictation or file transcription needs it instead of when EnviousWispr starts, and normally leaves memory after about 10 minutes without use. (v2.5.2)
• Help before you send feedback. When you press Send in Send Feedback, EnviousWispr shows up to three help articles that might answer your question. Read one, or send your message as you wrote it. On macOS 26 or later with Apple Intelligence on, you can mark each part of your message solved and choose not to send it. (v2.5.2)
• Shortcut warning in Keybinds. If macOS says a key combination is already taken, Settings > Keybinds shows a warning under that shortcut. (v2.5.2)
• Clipboard fix for Chrome and ChatGPT. Your clipboard comes back after you dictate into Chrome, Brave, Edge, Vivaldi, Arc or the ChatGPT app. (v2.5.2)
• Pastes go to the Chrome window you started in. Start dictating in one Chrome window, switch to another before you stop, and EnviousWispr goes back to the first window and pastes there. If you closed it, your words stay on the clipboard. (v2.5.2)
• More time to Undo a learned word. When Self-Learning saves a word, the Undo button stays for 4 seconds, with a bar showing the time left. (v2.5.2)
• Smarter Self-Learning. The Self-Learning Dictionary no longer learns edits that only add punctuation, such as "e mail" to "e-mail", and needs to be more certain before it learns from an edit you may not have finished. (v2.5.2)
• Introducing the Self-Learning Dictionary. When you correct a dictation, EnviousWispr detects the change and adds the intended term to your dictionary. (v2.5.1)
• Now available natively in German. Choose Deutsch in Settings > Appearance > Language, then relaunch to apply. (v2.5.1)
• Paste Last Dictation. ⌃⌘V drops your most recent transcription wherever your cursor is and ⌃⌘C copies it; Paste Last Dictation is in the menu bar too, and both shortcuts can be changed in Settings > Keybinds. (v2.5.1)
• Better at remembering the text box. EnviousWispr is better at pasting where it is supposed to paste. (v2.5.1)
• Introducing Send Feedback. Click the bug button beside Record to tell us what works and what doesn't, with an optional email if you'd like a reply. Your draft is saved on your Mac as you type. (v2.5.1)
• Apple Intelligence polish is faster and cleaner. On macOS 26 and 27 it finishes sooner, lays out spoken lists as lists and resolves more of your spoken corrections. (v2.5.1)
• Lower, mute or pause audio while dictating. Choose what your music and videos do while you dictate: Continue, Lower, Mute or Pause, on the Microphone page in Settings. (v2.5.1)
• Import your snippets. Bring snippets in from an exported file, a CSV, a pasted list, or straight from Wispr Flow or TypeWhisper, and review the list before anything is saved. (v2.5.1)
• Dynamic variables in snippets. Insert the current date, current time or your last copied text into a snippet. (v2.5.1)
• English (UK) spelling. Choose English (UK) as your dictation language and your text comes out in British spelling, like colour, organisation and centre. (v2.5.1)
• A Dock icon, and a window that opens in front. EnviousWispr appears in your Dock and the ⌘⇥ app switcher; turn off Show app in Dock under Settings > Appearance to keep it in the menu bar only. (v2.5.1)
• Addresses, links and codes in eight languages. With your dictation language set to French, Spanish, Polish, Dutch, German, Russian, Portuguese or Italian, spoken URLs, emails and technical codes come out written. (v2.5.1)
• Versions, IP addresses and codes come out whole. "version two point five point zero" becomes version 2.5.0 and "localhost colon three thousand" becomes localhost:3000. (v2.5.1)
• Better support for /commands. Spoken slash commands such as "slash clear" format as /clear, even with Spoken punctuation turned off. (v2.5.1)
• Mouse wheel scrolling glides. The mouse wheel moves farther per notch with a short smooth glide; trackpad scrolling is unchanged. (v2.5.1)
• Introducing Transcribe a File. Drop in a voice memo, lecture, meeting or video you already have and get back clean text, with speaker turns when the app can tell the voices apart, and, with a polisher selected, spoken lists laid out as real lists and the words the polisher removed, replaced or added marked up so you can see them. It takes one file at a time, transcribes it on your Mac, and the recording never leaves it; polish follows the polisher you pick, on-device by default. (v2.5.0)
• Apple Intelligence polish improved on macOS 27. On macOS 27, Apple Intelligence now removes the opening ums, lays out spoken lists and resolves more of your self-corrections, and long dictations that fit its larger model are no longer skipped. The Apple Intelligence card in Settings shows which model is running and its capacity on both versions. On macOS 26 the polishing itself is unchanged. (v2.5.0)
• Improved capitalization when dictating into an existing sentence. When you dictate after words you already typed, EnviousWispr matches your capitalization. If the app was busy, that matching could switch itself off until you relaunched; it now stays on in far more cases. (v2.5.0)
• Dictated measurements keep their unit. Say "the gap is 5 mm" or "120 mm Hg" and the unit comes through with the number. (v2.5.0)
• No more freeze after sitting idle. The speech engine now runs directly inside EnviousWispr instead of a separate helper that macOS could shut down behind its back, so the silent few-second freeze on the first press after idling is gone. (v2.5.0)
• Spoken email addresses convert in more languages' own words. While dictating in English, an address spoken with another language's words for at and dot, such as "marco arroba esempio punto com", comes back as [email protected], and addresses ending in a country domain such as .de, .nl or .fr convert too. (v2.5.0)
• The recording pill honours Reduce Motion. With Reduce Motion on in macOS, the pill no longer breathes, flashes or pulses; the colours stay so a warning still reads as a warning. (v2.5.0)
• History tells a dictation from a transcript. Each row says which kind it is, and All, Dictations and Transcripts buttons above the list show one kind or both. (v2.5.0)
• Eight small fixes across dictation, paste and History, including a cancel that stops waiting on a stalled transcription and keeps your audio for recovery on the next launch, digit-by-digit decimals that format as one number, and a paste that no longer waits on an unresponsive target app. (v2.5.0)
• A second on-device polish model, S1-mini by Superwhisper. A small open model for cleaning up dictation, now an option beside EG-1 in AI Polish settings. A 484 MB download, runs on your Mac, and free. Three writing style settings, Tone, Structure and Context, let you choose how it writes. Happiest in English. EG-1 stays the recommended choice. (v2.4.7)
• Snippets. Say a keyword, then a short phrase, and the text you saved is pasted word for word. A fresh install starts with six working examples to try. (v2.4.7)
• More of EnviousWispr is switched on from the start. Escape Recovery, Live Preview and the recording sounds are now on by default, and each is a switch you can turn off. Quick Add has moved to Control Shift W so it no longer shares a key with recording. (v2.4.7)
• Add a custom word to your dictionary with a click of a button. Highlight a misheard word anywhere on macOS, press the Quick Add keybind, and save the right spelling without opening Settings. A small panel ranks it against the words you have already saved, so one Return usually finishes the job. It is also in the menu bar menu. (v2.4.6)
• A new pill design for recording, and a practice run during setup. Choose between recording pill designs in Appearance, and try dictation before onboarding ends. (v2.4.6)
• Dictated web addresses come out ready to use. Stray periods and dot words around a spoken address are cleaned up, a garbled https is repaired before polish sees it, and pasting into a browser's address bar no longer adds a trailing space. (v2.4.6)
• Real German, Dutch, Danish and Norwegian words are no longer deleted. Filler removal now reads the language you are dictating in, so "er" and "um" survive where they are real words. (v2.4.6)
• Recover a dictation you cancelled by mistake. Escape Recovery is on by default, so hitting your cancel keybind keeps the text for 24 hours instead of throwing it away. The Cancel button still discards deliberately. Only the transcript is kept, never the audio. (v2.4.5)
• See your words as you speak. The recording pill can now show what it is hearing, live, with a choice of two preview engines. The text you actually get is unchanged: it is still transcribed from the whole recording when you stop. (v2.4.5)
• Dictation no longer records silence on a virtual microphone. If your Mac's default input was a virtual device (Krisp, Loopback, BlackHole, an aggregate, a meeting app's mic), the app bound it and captured nothing at all. It now prefers a real microphone. (v2.4.5)
• EG-1 retrained on its weakest spots. Announcing a list out loud produces a real list far more often, and changes of mind mid-sentence now land level with a frontier cloud model on our held-out benchmark. It is also about 14% faster. (v2.4.5)
• Bring your vocabulary over from another dictation app. Custom words import from eight other Mac dictation apps, from a file, or from a pasted list, and export to a backup you own. (v2.4.1, expanded in v2.4.5)
• Use the Globe key for dictation. The Globe key, marked Fn on many Macs, works as your keybind. (v2.4.4)
• Bluetooth and USB audio properly supported, with customizable sound cues and your choice of recording pill location. (v2.4.0)
• Meet EG-1, our own on-device polish model, and a full visual refresh of Settings. (v2.3.0)
• Dark mode, and recordings up to an hour long. (v2.2.0)

Quick Start

Install with Homebrew:

brew install --cask saurabhav88/tap/enviouswispr

Or download manually:

1. Download EnviousWispr.dmg from the latest release
2. Drag to Applications, launch
3. Grant Microphone, Accessibility, and (on first paste fallback) Automation permissions when prompted
4. Set your preferred keybind on the Keybinds page
5. Start talking

Requirements

• macOS 14 (Sonoma) or later
• Apple Silicon (M1 or newer)

Building from Source

git clone https://github.com/saurabhav88/EnviousWispr.git
cd EnviousWispr
swift build            # compiles the Swift packages (dependencies resolve via SPM)

The runnable .app is assembled by the Xcode build engine via Tuist, not by swift build. Use ./scripts/build-dev-app.sh for a local dev build, or the release path below. First build takes several minutes as ML models compile.

For a distributable .app bundle and DMG:

./scripts/build-release-dmg.sh <version>

The release build runs on the Xcode engine via Tuist, so it requires full Xcode (26+) plus mise and Tuist; set CODESIGN_IDENTITY to sign. Running the app itself requires macOS 14+.

Architecture

The app follows a pipeline state machine: idle --> recording --> transcribing --> polishing --> complete.

Key design choices:

• Swift 6 strict concurrency with full actor isolation
• Dual pipeline architecture with deliberately separate Parakeet and Whisper backends (isolation is a feature, not tech debt)
• Heart & Limbs pattern where the critical path (audio, ASR, paste) never fails, and features (polish, custom words, filler removal) degrade gracefully
• Local-first with LLM polish as an opt-in enhancement using your own keys

Contributing

Contributions are welcome. EnviousWispr is open source under the GPLv3. Please open an issue to discuss significant changes before submitting a PR.

This project uses conventional commits: feat(scope):, fix(scope):, refactor(scope):.

Privacy

EnviousWispr is built on a simple principle: your voice is yours.

• Audio is processed locally and is not uploaded. Local recovery can temporarily retain audio.
• LLM polish (if enabled) can run entirely on your Mac with EG-1 (our own model), Apple Intelligence, or a local Ollama model, so the polish step makes no network call. If you pick OpenAI, Gemini, or Claude, only text is sent (your transcript plus the polish instructions) using your own API key. If you pick a hosted Ollama model, the same text is sent to Ollama using your Ollama sign-in. Audio is never sent.
• Anonymous product analytics (PostHog) and crash reporting (Sentry) are on by default, and each can be turned off in App Settings > Privacy. A crash-reporting change applies after the app restarts.
• Crash reports exclude dictated text and audio. Technical context and identifiers are described in What Data Is Collected.

Connect

• Website: enviouswispr.com
• X: @EnviousLabs
• Email: [email protected]

License

EnviousWispr is open source under the GNU General Public License v3 (GPLv3), an OSI-approved license. You can read, build, modify, and redistribute the code under the terms of the GPL, including for commercial purposes; distributed derivative works must also be licensed under the GPLv3 with their source available.

Copyright (C) 2024-2026 Envious Labs LLC.

The EG-1 model is not open source. The GPLv3 covers the EnviousWispr application code only. EG-1's model weights are not part of this repository; they download separately and are distributed under the EG-1 Community Model License. You are free to download and use EG-1 within EnviousWispr, but you may not redistribute, re-host, mirror, or bundle the model into other products, or use the model, or outputs generated by it at scale, to train, fine-tune, or distill another model intended for commercial distribution. EG-1 is a fine-tuned derivative of Qwen3-4B-Instruct-2507 (Apache-2.0); this license applies to the fine-tuned weights.

The EnviousWispr name and logo are trademarks of Envious Labs and are not covered by the GPL.