Private AI workspace for Mac power users
Helping people work smarter at
Everything you need to run AI on your Mac
Best-in-class native macOS app: built using SwiftUI and AppKit.
Familiar UI, Universal Shortcuts Keys, and super-fast Apple M1/M2/M3/M4 performance.

- Your entire AI stack in one app.
- Use OpenAI, Anthropic, Google, Mistral, Azure, Bedrock, and local models in one unified workspace. Switch instantly without touching another app or browser tab.
- Serious workflow tools for power users.
- Work with projects, multi-chat threads, forking, and reusable agents to keep everything structured. BoltAI scales from quick prompts to complex long-running workflows.
- Multimodal intelligence built in.
- Analyze PDFs, screenshots, code, UI captures, and research documents with vision-enabled models. Get clear explanations, summaries, and structured results instantly.
- Fine-grained control over every generation.
- Tune temperature, max tokens, top-p/top-k, penalties, and your system instructions with precision. Get exactly the style and behavior you want from any model.
- Extend with tools, skills, and custom knowledge.
- Use code execution, MCP tools, and attached documents to build richer agents. Automate tasks, generate documents, or extract data without leaving the app.
- Privacy you can trust.
- Chats are stored locally and never used for training, and API keys can be encrypted with your passphrase. BoltAI keeps your work private, secure, and fully under your control.
Work faster anywhere
Bring AI to whatever you’re doing with a single shortcut. BoltAI pops up instantly, ready to explain code, rewrite text, summarize a webpage, or help with whatever’s on your screen — no context switching.
Apple intelligence, done right.

- Global shortcut.
- Open BoltAI instantly over any app with a single keystroke — no context switching, no friction.
- Screenshot → Answer.
- Capture your screen and get explanations, fixes, or summaries in seconds.
- Dictation & inline editing.
- Speak your requests or edit text on the fly for effortless, hands-free productivity.
Everything in one AI workspace
On all your devices
Use BoltAI on your Mac, iPhone, and iPad with the same chats, agents, and projects synced securely through the cloud.

Local models, zero latency
Run powerful LLMs locally via Ollama and LM Studio for fast, private responses without per-token costs.

Privacy built in
Chats are stored locally, API keys can be encrypted with your passphrase, and nothing is ever used to train models.

Extend with tools & MCP
Connect BoltAI to local and remote MCP servers, toggle tools per agent, and keep your AI workflows configurable in code.
{
"mcpServers" : {
"filesystem" : {
"args" : [
"-y",
"@modelcontextprotocol/server-filesystem",
"~/Data"
],
"command" : "npx"
}
}
}“My company provides critical software services for customers such as Spotify, Google, Coinbase, Binance, and many others. We use AI in our workflows (custom developed as well as LLMs) but there was no tool before BoltAI that tightly integrated LLMs into my workflow. I always had to click out, leave my current task, go to ChatGPT or similar and come back which broke my flow. BoltAI makes everything effortless. It's a superpower.”

A powerful workspace for serious users
BoltAI brings together pro tools, advanced models, structured workflows, and deep customization — all in a clean, native Mac experience built for speed.
- Advanced agents
- Build reusable agents with custom instructions, knowledge, and tools for consistent high-quality outputs.
- Context profiles
- Switch between tailored configurations for coding, writing, research, or analysis instantly.
- Multimodal analysis
- Work with screenshots, PDFs, and documents. BoltAI extracts structure, summarizes, and explains.
- Multiple model providers
- Use OpenAI, Anthropic, Google, Mistral, Azure, Bedrock, and local models in one seamless workflow.
- Fork & branch chats
- Explore alternative ideas without losing your original thread. Perfect for comparison and iteration.
- Prompt library
- Save your best prompts once and reuse them anywhere. Build a personal toolkit for recurring tasks.
- Attachment-driven context
- Attach files, research, or reference documents. BoltAI manages context automatically and intelligently.
- Advanced LLM controls
- Fine-tune temperature, token limits, top-p/top-k, and penalties for precise, predictable results.
Latest release
BoltAI 2.18.5
5 days ago
Build 97This update improves how chat history and attachments are sent to the model.
Improved
- Improved how chat history is sent to the model in long chats.
- Chat Context now defaults to the last 20 message pairs instead of 8. Chats, projects and agents with their own setting keep it.
- In the Inspector, settings that come from a project, an agent or your defaults are now shown read-only. Click Edit… to customize the chat or change the settings where they come from.
- Long chats that send all messages now show a notice above the composer, with an option to send only the last 20 pairs.
- Large images are scaled down to fit provider limits when sent, in their original format. The attached file isn't changed.
Fixed
- Fixed Chat Context applying only to the first request when the model uses tools. Later steps sent the full chat.
- Fixed attachments the model can't read being dropped silently or sent as files the provider rejects. The model is now told which file it couldn't read.
- Fixed an image that couldn't be loaded being left out of your message. The send now stops and tells you.
- Fixed spreadsheet attachments reaching the model without the values Excel shows, such as dates and percentages.
- Fixed RTF attachments being sent as RTF markup instead of text.
- Fixed Quick Compose and the Instant Chat Bar dropping your default system prompt and parameters when you picked a model.
- Fixed agents set to "Use Default" Chat Context ignoring your account's Chat Context.
- Fixed a reply you stopped being sent back to the model with its unfinished reasoning. Replies cut off by the output limit keep their reasoning.
- Fixed Save & Regenerate on an earlier message regenerating the wrong reply while the chat showed an error.

