AlternativesMac6 min read
Open source Superwhisper alternative for Mac
Want an open source Superwhisper alternative? Petal is free and MIT licensed, with local models, your own cloud keys, and a per-app Router at no cost.
Published by Optimal Apps, the maker of Petal.
On this page
The main open source Superwhisper alternative for Mac is Petal. Superwhisper has a good free tier with unlimited local Whisper models, but cloud models, your own API keys, and file transcription need Pro, at $8.49 a month or $249.99 for a lifetime license. Petal is free and MIT licensed. It runs speech recognition and the Petal W1 cleanup model on your Mac, and it lets you add your own OpenAI, Anthropic, OpenRouter, Ollama, or LM Studio connection, with a per-app Router, at no cost. Optimal Apps, the publisher of this site, makes Petal.
Superwhisper vs Petal at a glance
| Superwhisper | Petal for Mac | |
|---|---|---|
| Price | Free; Pro $8.49/month, $84.99/year, or $249.99 lifetime | Free |
| Source code | Closed | Open source, MIT |
| Local speech models | Free plan: unlimited Whisper models | Parakeet, Whisper Large V3 Turbo, Qwen3 ASR, Voxtral Realtime, Apple Speech |
| Local AI cleanup | Custom prompts on the free plan | Petal W1 on device, or Apple Intelligence |
| Your own API keys | Pro | Free |
| Cloud models | Pro | Free with your own key |
| Per-app behavior | Modes | Router: a route for each app or website |
| Audio file transcription | Pro | Free: drop a file on the menu bar icon |
| Platforms | macOS, Windows, iOS, Android | macOS; Windows as a separate project |
| Intel Macs | Cloud models recommended | Apple silicon needed for local models |
Superwhisper prices are from its pricing page on 20 November 2026. Superwhisper offers 40% off for students and a 14-day refund on all plans.
What Superwhisper's free plan includes
Superwhisper's free plan is generous, and it is fair to say so. It includes:
- Voice to text in any app
- Unlimited use of local Whisper models
- Custom prompt control
- More than 100 languages
- Meetings (Mac only)
Pro adds your own AI API keys, unlimited cloud and local AI models, translation to English, transcription of audio and video files, and priority support. You can try Pro free for 3,000 words.
So the question is not "free or paid" in general. It is which features you need. If you only want local Whisper dictation, Superwhisper's free plan can be enough.
Why look for an open source Superwhisper alternative
You want your own API key without a subscription
Many people already pay for an OpenAI or Anthropic account. In Superwhisper, using that key is a Pro feature. You pay Superwhisper on top of what you pay the provider.
In Petal, your own key is free:
- Open Settings > Intelligence and choose Cloud Model.
- Pick OpenAI, Anthropic, or OpenRouter and paste your key, or point Petal at an Ollama or LM Studio server.
- Choose a model and a preset: Clean Up, Email, Notes, Professional, AI Prompt, or Assistant.
- Test the prompt in Try It before you dictate.
Petal stores the key in the macOS keychain. Only the transcript goes to the provider you chose. With Ollama or LM Studio on the same Mac, nothing leaves the computer.
You want the code to be open
Superwhisper's code is closed. Petal's full source is on GitHub under the MIT license. You can read how it records, what it sends to a cloud provider, and where it saves history. You can also build it yourself or change it.
The lifetime price
$249.99 is a fair price for a tool that you use all day, but it is a lot to pay up front to try cloud features. Petal has no paid tier on the Mac, so there is no lifetime decision to make.
Setup time
Third-party reviews often say that Superwhisper takes time to set up: models, permissions, and modes. Petal's first run is one guided setup: pick a speech model, set a shortcut, allow Microphone and Accessibility access, and choose a cleanup option.
Petal's Router vs Superwhisper modes
Superwhisper uses modes, such as a mode for email or a mode for chat, each with its own prompt. Petal uses a Router.
In Settings > Router, you add a route for an app or a website. Each route can use its own intelligence:
- Petal W1 for fast on-device cleanup
- Apple Intelligence
- Cloud Model, with the provider, model, and tools from Settings > Intelligence
- A preset prompt, a custom prompt, or As Said, which skips cleanup and pastes your words
For example, you can paste raw text in Terminal, use Petal W1 in Slack, use the Email preset in Mail, and send prompts to the AI Prompt preset in a chat site. Petal picks the route from the front app, so you do not switch modes by hand. Read per-app dictation cleanup on Mac for a full setup.
Paste behavior
A common Superwhisper complaint is that the old clipboard pastes instead of the dictation, or the text pastes twice. Superwhisper's docs fix this with its Clipboard content and Simulate keypresses settings. See Superwhisper pastes old clipboard and Superwhisper not pasting if you stay with it.
Petal handles this with a delay. With Restore Clipboard on in Settings > Recording, Petal waits 700 ms after the paste before it puts your old clipboard back. Electron apps such as VS Code and Slack read the clipboard late, and the delay gives them time to read the dictation first.
What leaves your Mac
Both apps can run fully on device, so the question is what happens when you turn on cloud features.
| Setup | Superwhisper | Petal |
|---|---|---|
| Local speech, no AI | Stays on the Mac | Stays on the Mac |
| Local speech, local cleanup | Stays on the Mac with a local model | Stays on the Mac with Petal W1 or Apple Intelligence |
| Cloud model | Pro; audio or text goes to the provider | Text goes only to the provider you pick |
| Local server (Ollama, LM Studio) | Check Superwhisper's docs; own keys need Pro | Free; text stays on the Mac when the server runs locally |
Petal never sends your audio to a cloud cleanup provider. It transcribes on your Mac first, then sends only the transcript. If you turn on Screen for a cloud route, Petal also sends a screenshot, so the model can spell names on screen. Screen needs Screen Recording access and is off until you turn it on.
Third-party reviews note that Superwhisper saves recordings to disk by default. In Petal, you choose what history keeps in Settings > History: transcripts, audio, both, or nothing.
Where Superwhisper is ahead
- More platforms. Superwhisper runs on macOS, Windows, iOS, and Android from one company.
- Meetings. Superwhisper records meetings on the Mac. Petal is for dictation and dropped audio files, not meeting capture.
- Translation. Pro translates from any language to English.
- Intel Macs. Superwhisper Pro cloud models run well on Intel. Petal's local models need Apple silicon.
- Enterprise. Superwhisper offers an enterprise plan with SOC 2 Type II.
How to try Petal next to Superwhisper
- Download Petal from the Petal for Mac page.
- Open it and follow the setup. Pick Parakeet TDT 0.6B V3 for fast English, or Whisper Large V3 Turbo if you are used to Whisper.
- Set a shortcut that does not conflict with Superwhisper's. Superwhisper's default is Option-Space.
- In Settings > Intelligence, choose Petal W1.
- Dictate the same paragraph with both apps into the same app, and compare.
Summary
Superwhisper's free plan is a good local Whisper tool. Pick Pro if you want its cloud models, translation, and meetings on several platforms. If you want an open source Superwhisper alternative with your own keys, cloud cleanup, a per-app Router, and file transcription for free, try Petal. For a detailed comparison, read Petal vs Superwhisper or Wispr Flow vs Superwhisper.
Questions
Is there an open source Superwhisper alternative?
Yes. Petal for Mac is open source under the MIT license. It runs speech models on your Mac, cleans up text with the on-device Petal W1 model, and supports your own cloud keys at no cost.
How much does Superwhisper Pro cost?
On 20 November 2026, Superwhisper's pricing toggle showed Pro at $8.49 per month, $84.99 per year, or $249.99 for a lifetime license, with 40% off for students. The free plan includes unlimited local Whisper models.
Can I use my own API key for free in Superwhisper?
No. Superwhisper lists 'use your own AI API keys' as a Pro feature. Petal lets you add your own OpenAI, Anthropic, or OpenRouter key, or a local Ollama or LM Studio server, for free.
Does Petal work on Intel Macs like Superwhisper cloud mode?
Petal's local models need an Apple silicon Mac. Superwhisper's FAQ says Intel Macs work best with its cloud models, which need Pro.