Mac-first knowledge workers
Improve email, Slack, notes, tickets, and documents without repeatedly copying work into a standalone chatbot.
Independent tool overview
Kerlig at a glance
Kerlig is a native macOS AI assistant that brings writing, chat, document analysis, and reusable actions into almost any Mac app. It uses your own provider API keys or supported local models and charges a one-time app license instead of a subscription.
Visit the official Kerlig site ↗
- Best for
- Mac users rewriting, summarizing, researching, and chatting with content across apps
- Platform
- macOS 12+ on Apple Silicon and Intel Macs
- Model access
- 350+ hosted models with your API keys, plus supported local models
- App pricing
- One-time licenses currently start at $49, plus any provider usage
- Data path
- Direct to the selected provider; chat history is encrypted locally
Overview
What Kerlig is
Kerlig is designed to reduce context switching. Select text in Mail, Slack, Notes, Notion, a browser, or another supported app, press a system-wide hotkey, and run an AI action without moving the work into a separate browser tab.
The app supports hundreds of models through OpenAI, Anthropic, Google, Groq, OpenRouter, and other integrations, plus on-device Apple Foundation Models and local Ollama or LM Studio models. Model usage is paid to the provider unless it runs locally.
Kerlig is most useful for people who want a configurable Mac-native AI layer. It is not a bundled unlimited AI service, a Windows app, or a team knowledge platform, and using third-party models still sends selected content to those providers.
Use cases
Who Kerlig is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Multi-model users
Switch between provider and local models based on quality, privacy, speed, context, or cost.
Privacy-conscious individuals
Keep chat history on the Mac and use Apple Foundation Models, Ollama, or LM Studio when content should not leave the device.
Teams standardizing Mac actions
Deploy licenses and supported provider keys through command-line tooling for repeatable Mac setups.
Capabilities
Core Kerlig features
Use AI in other Mac apps
Capture selected text or approved app context from browsers, Mail, Slack, Notes, Notion, and other workflows when Kerlig is opened.
Built-in and custom actions
Run proofreading, rewriting, summarization, translation, extraction, or custom prompt workflows with system-wide hotkeys.
Headless actions
Run an action in the background, then place the result back into the source app or clipboard without showing the full Kerlig interface.
350+ model choices
Connect provider API keys and change the preferred model by task. Kerlig's August 2026 release refreshed support for current OpenAI, Anthropic, Google, Groq, and other models.
Local and on-device models
Use Apple Foundation Models on supported Macs or connect Ollama and LM Studio for offline, local inference.
Attachments and app context
Analyze documents, images, webpages, app content, and YouTube transcripts; supported captured content can also be saved as Markdown.
Presets, prompts, and variables
Create reusable tones, instructions, prompt templates, and dynamic variables such as selected text, current app, date, or prior response.
Local encrypted history
Chats are stored in an encrypted database on the Mac and can be revisited or cleared by the user.
Process
How the Kerlig workflow works
Step 1
Choose the data path
Use a local model for sensitive work or select a provider whose data terms meet your needs. Add the API key and set budgets directly with that provider.
Step 2
Create task-specific actions
Build separate actions for proofreading, executive summaries, customer replies, ticket cleanup, or research rather than relying on one vague prompt.
Step 3
Run from the source app
Select the relevant text, use the hotkey, confirm the captured context, and run the action with the model best suited to the task.
Step 4
Review before inserting
Check facts, tone, recipients, sensitive information, and unintended changes before pasting or sending generated text.
Cost
Kerlig pricing and free plan
Kerlig sells one-time app licenses. Hosted AI tokens are not included and are billed separately by the provider tied to your API key. Local models can avoid provider usage fees but require compatible hardware and setup.
Basic License
$49 one time, current sale
For one person using one Mac.
- One Mac and one seat
- Vision, attachments, presets, and tone settings
- 350+ hosted models with your own API keys
- Apple Foundation Models, Ollama, and LM Studio
- 48-hour email support
Pro License
$79 one time, current sale
For one user working across two Macs.
- Two Macs and two seats
- Same model and workflow features as Basic
- 24-hour email support
Team License
$297 one time
For small Mac fleets and remote setup.
- Ten Macs and ten seats
- Remote command-line deployment
- One year of app updates
- Priority Zoom and email support
- Contact Kerlig for additional seats
Additional AI usage
Provider or local-model cost
The license does not bundle hosted model tokens.
- OpenAI, Anthropic, Google, Groq, OpenRouter, and other providers bill under their own terms
- Local Apple, Ollama, or LM Studio models can run without per-token provider charges
- Kerlig offers a 14-day money-back guarantee
- VAT may apply
Pricing checked . Check current pricing at the source ↗
Assessment
Kerlig strengths and limitations
Where it stands out
- Native Mac workflow avoids repeated browser-tab switching
- Broad model choice prevents dependence on one AI provider
- One-time license can be economical for long-term use
- Local model options provide offline and more private workflows
- Highly configurable actions, hotkeys, prompts, variables, and tones
- App context, attachments, multi-turn chat, deep research, and citations cover more than simple proofreading
- Encrypted local chat history and direct provider connections reduce Kerlig's role in the data path
What to consider
- Mac only; it does not support Windows, Linux, iPhone, or Android
- Hosted model tokens are a separate recurring cost and require users to manage provider accounts, API keys, limits, and billing
- Using cloud models sends selected content to the chosen provider even though Kerlig does not store it on its own servers
- Capturing content from other apps requires macOS Accessibility and sometimes Automation or browser scripting permissions
- Local model quality, context, and speed depend heavily on the Mac and model selected
- The sale page's update-entitlement wording is inconsistent between its promotional label and plan bullets, so confirm whether a purchase includes one year or lifetime updates
Compare
Kerlig alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Business Operations
Raycast
Choose Raycast when you want a broader Mac launcher and automation platform with AI as one part of a larger productivity workflow.
Explore Raycast →Business Operations
ChatGPT
Choose ChatGPT when you prefer a bundled consumer subscription and do not want to manage separate provider API keys.
Explore ChatGPT →Project Management
Claude
Choose Claude when a dedicated first-party assistant, projects, artifacts, and Anthropic's own model experience matter more than system-wide Mac actions.
Explore Claude →Questions
Kerlig FAQs
What is Kerlig?
Kerlig is a native macOS AI assistant that can rewrite, summarize, research, and chat with selected content from other Mac apps using a system-wide hotkey.
How much does Kerlig cost?
Current one-time sale pricing starts at $49 for one Mac, $79 for two Macs, and $297 for ten Macs. Hosted model usage is paid separately to the AI provider.
Does Kerlig include ChatGPT or Claude usage?
No. Kerlig connects to supported providers with your own API keys, so provider usage and billing are separate. You can also use supported local models without hosted token charges.
Is Kerlig private?
Kerlig says chats are stored in an encrypted local database and cloud requests go directly to the selected AI provider. For content that must stay on-device, use a compatible Apple Foundation Model, Ollama, or LM Studio model.
Which Macs support Kerlig?
Kerlig supports Apple Silicon and Intel Macs running macOS 12 or later. Specific on-device models can require newer macOS versions and compatible hardware.
Can Kerlig work with text in other apps?
Yes. It can use selected text or approved structured context from supported browsers, Mail, Slack, Notes, Notion, and other apps after the required macOS permissions are granted.
Bottom line
Our Kerlig verdict
Kerlig is a strong fit for Mac users who want AI available at the point of work and prefer choosing their own cloud or local models. The one-time license is attractive, but buyers should budget for API usage, review app permissions, and confirm update entitlement before purchase. A bundled first-party assistant is simpler; Kerlig wins on Mac integration, model choice, and customization.
Visit Kerlig website ↗Similar Content Creator Tools
View all Content Creator AI tools →
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.