Mr Broccoli

Name: Mr Broccoli
Author: Tobias Winkler

Your keys. Your models.

0 ratings

📊 Productivity

🔍 Reference

Free

Rating summary

Details

Released
Updated
Version
Compatibility
Age Rating
Size
Devices
Data updated
Languages

August 11, 2026
September 28, 2026
4.2.0
iOS 16.4+
4+
56.87MB
iPhone, iPad
8 days ago
American English

Features

Developer

Tobias Winkler

✨Promote your app ✨

📝1+ tags

Tags↘︎

🔄6+ apps

Alternatives↘︎

📦~ in 3 weeks

Updates frequency↘︎

🖼️Get Icon

Icons↘︎

About

Mr Broccoli is a voice-first AI app for people who already have their own provider accounts. Add your keys, choose exactly which model answers, and listen to the reply. Nothing routes through a server of ours, because there isn't one. BEFORE YOU BUY Mr Broccoli is built around your own API keys from OpenAI, Anthropic, Google Gemini, xAI, Mistral, DeepSeek, OpenRouter or Alibaba Qwen. Your purchase gives you the app. It does not include AI usage — your provider bills you directly for what you use. TALK, DON'T TYPE • Speak, type or attach an image • Answers are read aloud, starting as soon as the first paragraph is ready • Skip by paragraph, pause, resume or restart playback • Drive Session for hands-free turns: it waits until you finish, counts down, sends, speaks the answer and listens again • Speech recognition using the system, an on-device model or your provider CHOOSE WHO ANSWERS • Save up to four routes — provider, exact model and reasoning effort • Switch between them mid-conversation without losing context • The home screen says who answers next and at what effort • Every answer records the route that actually produced it, including any fallback TUNE EACH CONVERSATION • Length: brief, normal or thorough • Tone: professional, casual, nerdy, concise, Socratic or ELI5 • Your own model and speech-delivery instructions • A specific voice per conversation Set defaults globally, override them per conversation. YOUR CONVERSATIONS • Kept in a local database on your device • Search, pin, archive, rename, share or delete • Fold and expand entries in one continuous transcript • Branch from any message to try another direction without losing the original • Lock individual conversations behind a password or biometrics MORE WHEN YOU WANT IT • Web search grounding, with sources attached to the answer • Council: several models answer, review each other, then one synthesises • Image prompts to compatible models • Optional recall from your past conversations, entirely on the device PRIVACY • No Mr Broccoli account and no Mr Broccoli server • Requests go straight from your device to the provider you chose • API keys stay in the operating system's secure storage — never in backups, logs or exports • A hosted provider still receives the request you send it YOURS TO KEEP • Full backups as readable JSON or passphrase-encrypted AES-256-GCM • Markdown archives to read anywhere or hand to another AI • Importing never overwrites what is already there 19 languages, including Arabic and Urdu right-to-left. Interface, dictation and answer language are set independently. Screen reader, large text, contrast and reduced-motion support throughout. WHAT IT DOESN'T DO No cloud account, no cross-device sync, no hosted inference, no included credits, no silent switching of your chosen model or voice, and no autonomous actions on your behalf. AI output is not medical, legal or financial advice. No subscription. Because there is no account system, a purchase on one store does not unlock the app on the other.

What's New in Mr Broccoli

4.2.0

September 28, 2026

Mr Broccoli 4.2 brings the newest AI models, better speech, and a long list of reliability fixes. NEW MODELS • GPT-6 Astra, GPT-6 Sol and GPT-6 Luna, with every supported reasoning effort up to Max. • Claude Opus 5.5, Claude Opus 5 and Claude Fable 5.1. • Grok 4.7 and Grok 4.6. • Gemini 3.8 Flash and Gemini 3.7 Flash. • Qwen 3.8 Max, Qwen 3.8 Flash and Qwen 3.7 Flash, with model-specific reasoning controls. • DeepSeek V4.1 Flash, which now also understands images. DeepSeek models add a Low reasoning setting. • OpenRouter adds current versions of GPT-6 Astra, Sol and Luna, Claude Opus 5.5, Opus 5 and Fable 5.1, Grok 4.7 and 4.6, Gemini 3.8 Flash, Qwen 3.8 Max, and DeepSeek V4 Pro and V4.1 Flash. SPEECH • GPT Transcribe is the new default for OpenAI speech input, and Gemini 3.5 Transcribe adds word-for-word transcription. • New Gemini 3.8 Flash and Flash-Lite voices for spoken replies. • Long dictation now keeps every sentence, not just the last one. • System Recognition can be chosen whenever iOS offers it, even when strict offline recognition is unavailable. • Full reply first pauses between paragraphs again, and cancelling reliably stops any queued audio. • Voice previews can now replace a paused spoken reply. • Voices saved for a conversation stay selected when a provider updates its models, and OpenAI's older voice models only show voices they support. WORKSPACE • A cleaner voice and text layout: the arrows, voice orb and composer sit on one line, Image and Hands free move to the edges, and Council and Web stay centred. The composer and send button stay visible on small screens and with the keyboard open. • Adding an answering model no longer leaves Settings unresponsive. RELIABILITY AND PRIVACY • Starting a second recording or Hands free no longer stops a recording already in progress. • Web search keeps usable results, rejects empty answers and cancels stalled responses. Gemini and xAI search no longer ask the provider to store your requests. • Model Council now credits the provider and model that actually answered. • Backup restores and conversation branches report failed saves instead of claiming success. • Locked conversations stay locked if the app goes to the background while one is loading. • The app recovers from temporary storage start-up problems without a restart. • A damaged settings file no longer hides your saved API keys. • Slow requests always end with a clear timeout message. CHANGES • Qwen speech input and output are removed ahead of Alibaba retiring them on October 10. Qwen chat and search are unchanged, and saved Qwen voices switch to your device's built-in speech. • Google now limits Gemini 2.5 to accounts that already used it, so it is no longer offered for new choices. Existing selections keep working, and older Gemini 2.5 voices move to Gemini 3.8 with the same voice. • OpenAI Realtime models are hidden until they are fully supported. Saved selections move to GPT-5.6 Sol or GPT-5.4 mini. • The ElevenLabs built-in fallback voice is now correctly named Janet.