You know how it goes. You are on a business trip, vacation, or conference, and you come back to an inbox so full you wish you had hired an assistant to sift through the messages and pick out the important follow-ups. Did you know Claude can do that for you with a single sentence prompt?

Recently, I celebrated my birthday, then traveled to a friend’s milestone birthday for a week. Once I got home, I was at the HumanX conference for another week. Every day, my inbox kept piling up.

AI’s superpower is cutting through the noise

Claude, ChatGPT, and Gemini are trained on human-generated content, so it is no surprise that they can read your inbox and find the important things.

When I sat in my chair this morning, I used a single sentence prompt to identify important follow-up items. Out of about 200 emails I have yet to read, Claude found about 20 that required a detailed review, filtered out 12 as noise, and identified the 8 that required my immediate attention.

I asked Claude to generate a chart showing how it filtered my emails, so this is a real snapshot from my Cambermast inbox:

The One-time setup that takes two minutes

As I mentioned, this Claude feature is built in, but you do need a paid account.

One-time setup: Connect Claude to your Gmail account

1. Go to https://claude.ai and log in.

2. Start a new chat (you do not want to do this in an existing chat, as Claude may not be as accurate)

3. In the prompt box, select the + item, then select Connectors, and finally select Gmail, as shown in the following image.
   💡 If this is your first time connecting to Gmail, select the Add connector item, select Gmail, then enable it.

3. In the same prompt box, select the model picker and then enable Extended thinking.
   💡 While Claude should be able to find your follow-up meetings with a basic prompt using the default model, I recommend you enable Extended thinking, as the results were better for me. You should try both and see what works best for you.

That’s it! Now you can begin finding follow-up emails.

The single-sentence prompt to get your inbox items

Once Gmail is connected, you can prompt with any number of questions, but the point of this email is to help you find follow-up items.

Prompt:

Please review my inbox and let me know if there are any items I should follow up on, especially immediately.

Claude did the rest. It searched through my unread messages, prioritized them by urgency, and gave me a clear action list.

Here are the (anonymized) results after my conference last week:

How it works and why it matters

Claude can access my email data, understand context, distinguish between signal and noise, and present information in a way that respects my time.

The email overload problem has not been solved since Gmail launched. We create folders, labels, filters, and rules, but those are all manual systems that require you to anticipate your own needs. An agent works differently. It understands what matters to you in real time and brings it to the forefront.

Important limitations

As of this writing, there are a few things you should know about Claude’s limitations.

Read more

You already have the data. It’s sitting in your inbox. In a few minutes, you can turn recent email conversations into a clean contact list, ready for Google Contacts.

Read more

Recently, I have been talking with people leading AI initiatives in companies. In early 2025, there were many failed efforts, with only a smattering of shining examples. Later in the year, I have been talking to people who have shared real examples of how AI Agents and automations are helping grow their business.

Read more

• Agents become your teammates in Slack

• AI in action: the next chapter for Jira Service Management

• Zoom’s “AI Companion” has been extended with agentic capabilities: it can connect to about 16 third-party tools

• Vibe working: Introducing Agent Mode and Office Agent in Microsoft 365 Copilot

But what are AI Agents and Agentic AI? In this article, I will provide you with some guidance on this emerging technology and how it could impact your business.

The birthplace of AI Agents

The “You are a…” prompt

The reason AI is so capable at understanding your prompt is that it was trained on a massive corpus of publicly available text or licensed data. Want trip planning advice? No problem. Want to learn quantum mechanics? Yup, it can do that too.

In ChatGPT’s early days, there were fun social media posts showing people’s “you are a…” prompts, like this:

Prompt: You are a salty pirate captain. Get me the weather and tell me how to dress.

AI Response: Arrr, the next two days in San Francisco will be sunny with cool mornings near 50 °F and afternoons risin’ to the upper 60s and low 70s. Wear layers, matey — a light shirt, warm sweater, and windbreaker with sturdy shoes to keep ye steady on the deck.

It's funny, sure. But early adopters figured out that this fun trick could measurably improve content with real-world business implications.

The idea is this:

AI understands your job, your motivations, and the best practices you follow. So it only makes sense that you can ask AI to become something you are not and ask for advice.

Imagine you are speaking at a conference. You are not great at self-promotion. You prompt AI to write a social media post and receive a response that seems reasonable. Or, you could use AI more strategically with the "You are a…" prompt to get a better response:

Since AI has been trained on the world’s knowledge, it knows how promoters and social media experts get people’s attention. That is why adding a simple “You are a [role]” at the beginning of your prompt yields much better results.

Introducing structured prompting

After a year or so of everyone experimenting with ChatGPT prompting techniques, a few patterns and best practices emerged, such as structured prompting.

In the following image, we are collaborating with AI to write marketing copy for our website:

• Knowledge provides a background to inform the AI response, like the copy for a product launch and our company’s brand identity.

• The prompt states the main request or task that you want the AI to complete, like creating marketing copy for a website.

• Instructions define the AI’s tone, style, and constraints to shape how it performs the task. Instructions can also define the AI's role, such as an SEO (search engine optimization) expert or a brand manager.

From instructions to Agentic AI and AI agents

Structured prompts are helpful, but typing them out over and over gets old fast. Worse, there is a lot of manual work you might need to do, such as:

1. Downloading the latest product and branding documents

2. Sending the document through the structured prompt

3. Uploading the document for a team review

4. Starting a Teams or Slack conversation to discuss the marketing copy

5. Scheduling a Zoom, Teams, or Google Meet call to get people on the same page

6. Send out an email informing people of the new marketing copy and when it goes live

That is where AI Agents come in. You can create agents that perform these workflows on your behalf. As you can see in the image below, I have a workflow agent that looks for new or changed products my company offers.

What I built is an Agentic AI workflow that coordinates two different AI agents to do work I used to do manually.

• An Agentic AI workflow can actively look for new or changed product documentation on your behalf. For example, it could look for a new document in Google Drive, OneDrive, Dropbox, responses to forms, or even check your email.

• A copy creator AI agent receives the product information, gets the latest brand and writing style guides, and generates the website copy.

• A copy coordinator AI agent will inform team members to review the material. The AI agent uploads the newly written copy from the copy creator agent to Google Drive and then sends a Slack message to the team with a link. It even sends out a Zoom link so everyone can review the product content together and send summary emails.

Now think about doing all that yourself every time. The meetings. The updates. The follow-ups. That is where AI agents can come in, and why you hear people make claims that they save 30% of their time or make teams 10x faster.

Making your apps more productive

Let's say you are in sales and use a CRM (customer relationship management) app. A potential lead goes to your website and fills out a form. Here is the typical data you receive, which is helpful but not necessarily actionable:

Now, let’s add a Customer Enrichment AI Agent that acts as your personal assistant, who finds out as much as they can about this person. Since this is for a business-related conversation, the AI agent looks up their LinkedIn profile, company data, and much more, like this:

Read more

I have been teasing AI agents for the past few weeks, and we are almost there! As you know, I like to be thorough, and there is one more topic we must cover together: the cost of running AI agents.

Before we build, let’s make sure your toolkit is ready

I cannot possibly provide you with a comprehensive list of every possible way to create agents. However, I can tell you that there is one tool I use almost exclusively because it is so well-known in the Agentic AI world: n8n.

There are other great choices, such as Zapier and Make. For those of you on Microsoft 365, you have access to a similar technology called Power Automate. A new kid on the block is Postman’s AI Agent Builder.

Here is why n8n stands out for me:

• The company offers an open-source version, allowing you to host your flows locally on your computer or corporate IT infrastructure.

• They offer paid plans on their website for those who prefer not to use SaaS (Software as a Service) with predictable pricing.

• They have adopted the popular open-source approaches to AI workflows, which are very popular among software developers (such as Langchain).

Why is that last point so important? Imagine you’ve built an automation that organizes customer support requests. You realize you could package it up and sell it as a product. You can do that with n8n!

The new breed of AI agent builders

You may be aware that Anthropic’s Claude can create apps in a chat window using their Artifacts feature. Other products can make a full-blown app for you from a prompt, such as Loveable or Replit.

Even developers can create agents using code editors like GitHub Copilot, WindSurf, and Cursor.

All these products offer the promise of quickly transitioning from an idea to a product, with minimal knowledge of the underlying technology used to run them.

Unfortunately, the underpinnings of some products are so opaque that you don’t really know how they work.

When I review automation agents, I test them across different platforms before settling on the one that best fits the task. I prefer tools that let me visualize each step, since seeing the flow helps me understand exactly what is happening. That is why I often choose n8n, though I still use Zapier, Make, Power Automate, and other automation tools.

Clarifying prescriptive workflows vs AI automations

Imagine the following repetitive tasks where you and your team must send out a document to a large audience:

• Write a document.

• Send it out for review.

• Collect feedback.

• Revise and publish it.

A prescriptive workflow might:

• Send an email or Slack message to reviewers.

• Create and assign review tasks in a tool like Jira.

• Notify you when all tasks are complete so you know the document is ready.

A value-added AI automation could go further by:

• Help you draft the document.

• Use past feedback from your team and suggest edits.

• Assign reviewers and highlight topics that match their expertise.

• Rewrite the document based on input from your team and send you a final draft for approval.

• Publish the document, saving it to your repository, converting it to PDF, and even translating it.

The difference is clear. A prescriptive workflow is mostly an if-this, then-that process. An AI automation acts like a partner. It co-writes, suggests improvements, tailors tasks to the right people, and cuts down the time spent on manual edits.

Got it, but what about costs?

In general, you should expect to pay out-of-pocket expenses for the following:

1. Hosting your AI solution somewhere (whether you pay n8n or host it yourself). Call that $20/month.

2. AI API calls (read on for more on this important topic).

3. If your automation requires access to other tools, some are available for free, while others have a cost associated with them. If you want to perform basic automations with tools like Microsoft Office or Google Workspace, these will generally be free to use.

Aside from hosting costs, your most significant expense will likely be those AI API calls. Let’s understand what that means for you.

AI API tokens

AI tools use what we call tokens to define how much you pay. For example:

• 1 token ≈ 4 characters of English text

• A word averages about 1.3 tokens

Token pricing

Now that you understand the basics of what a token is, you should know there are typically two ways you will pay for them:

1. Input tokens represent your prompt (instructions) that define what you want from the AI.

2. Output tokens represent the AI’s response to your prompt.

Input tokens usually cost less. Output tokens typically cost more. That is because input tokens are your instructions, while output tokens represent the AI generating a detailed response. The added cost reflects the value of producing new and valuable content.

How to estimate token cost

First, a big pricing disclaimer

For this article, I outline how to estimate pricing and provide some estimates in $ USD. Your pricing can vary widely based on usage, contracts, and any number of other factors. Do not use this article as a way to estimate your AI automation costs.

Take the time to check current rates and compare options so you get the best results at a cost that meets your budget. To explore pricing models directly, visit:

• OpenAI’s cost comparison site

• Anthropic’s pricing page

• Google’s Gemini developer API pricing page

Read more

Some people were building products, others were there to learn, and still others were presenting their work. I was there for all three.

I met many…

Read more

Article complexity: Technical, but not terrifying.

What are AI agents and agentic workflows?

Workflows can mean almost anything in business. Traditional automation workflows follow fixed rules like a script. Agentic workflows use AI. They can adapt, make decisions, and hold conversations with people and other agents.

Traditional workflows

Let’s say you have a website where a customer fills out a form for technical support. It might contain fields like:

• Name

• Email address

• Priority (Low, Medium, High)

• Description of the issue

A traditional workflow always knows those fields will be required and can be run with a specific set of unchanging rules. If it sees “High,” it could send an email to an escalation manager for immediate response. A medium might follow another process, and a low priority follows yet another method.

A traditional workflow is a repeatable automation that always follows the same business rules and will reach the expected outcome every time.

Agentic workflows

When we introduce AI into these workflows, we move beyond simple rule-based automation. AI agents can make decisions, adapt to changing inputs, and even collaborate with other agents or people. Agentic workflows are the structures that connect these agents into a sequence of actions, where each step may involve reasoning, learning, or interaction, rather than just repeating the same rule every time.

While an agentic workflow can follow a set of rules, you should design it to be dynamic and adaptive. For example, imagine you run sales operations and the team likes to share a report with customers. Each salesperson, however, requests slight modifications, such as different date ranges, regions, or product categories. Every time, you have to stop what you’re doing, run the report, make the adjustments, add it to a sales deck, and send it back.

Guaranteeing some level of automation without AI can be very difficult. You would probably have to send the salespeople to a form that asks specific questions, and then the automation kicks in.

I can only imagine a flustered salesperson preparing for a big customer meeting. They will skip the form and send you a priority email. Now the automation is broken, and you are back to your old ways of getting that report manually.

Using AI, on the other hand, may be a pretty straightforward process, like this:

• An AI agent reads your emails.

• When it sees an email from a colleague asking for a unique report, the AI runs the report with the requested parameters.

• Once the report is complete, it generates the sales deck, drafts the email, and then sends it over to you for final approval (or sends the email if you are feeling so bold).

Pretty cool, right? It turns out that kind of workflow is super simple to create, and you can even do that yourself. But if you do want to develop personal AI automations, you will need to know a little jargon.

How computers talk to computers

I am talking to you with words, which is our shared language. Computers do the same thing, only their secret language is called APIs (Application Programming Interfaces). When two systems want to pass information back and forth, APIs carry the messages in structured sentences of data.

RESTful APIs: The messengers

Read more

Read more

What is ChatGPT Agent?

ChatGPT Agent is a capability that enables ChatGPT to act autonomously on your behalf by performing tasks such as browsing websites, filling out forms, running code, and more. ChatGPT Agent will perform a real-world execution of your request in a unified workflow that stays under your supervision.

That “under your supervision” part is super important. In the sample prompt I share later in this article, ChatGPT will require you to sign in to LinkedIn. That means ChatGPT Agent will stop and require you to intervene when it is time to enter a username and password, provide your credit card details, or provide personal information.

Here is how ChatGPT Agent works technically:

• It acts like a person who would go to the web and click through a browser to perform tasks on your behalf.

• The Agent runs within a secure virtual computer environment that includes a browser, a terminal for code execution, and connectors to apps such as Google Drive, Gmail, GitHub, and calendars (if you have opted to connect these tools).

• It utilizes real-time safety monitoring, watch-mode for sensitive tasks, disabled memory, and prompt-injection protection, and requires explicit user confirmation before executing actions such as purchases or sending messages.

Availability

As of this writing, ChatGPT Agent is available for the following plans:

• ChatGPT Plus

• ChatGPT Team

• ChatGPT Pro

Note: ChatGPT Agent is still rolling out for those accounts. If you don't see the feature today, it may appear tomorrow.

ChatGPT Enterprise and Educational accounts will receive Agent mode at a later date.

What can I use Agent for?

As a tech company, OpenAI touts the ultimate experiences of having ChatGPT plan a trip, a wedding, or a date night. While all that might be fine, I think business use cases are where the money is at.

Here are a few good examples:

• Lead list generation for potential customers or partners

• Run a competitive analysis based on your website and your competitors

• Review your LinkedIn profile to attract more people who may want to hire you

• Research venues for your upcoming tradeshow event

• Gather RFP examples and best practices from government or corporate websites

The secret to your first Agent's success: Turn-based prompting

If you have been reading my articles, you know I love the idea of prompting for a prompt. If you can think of a prompt, you can ask the AI to improve it on your behalf. For the prompt I share in this article, I will teach you how to create a turn-based prompt that will help you make valuable decisions about your LinkedIn profile.

Got it, Bill. Let’s see the prompt!

Create a new ChatGPT 4o chat and paste the following prompt:

Read more

Can I create a recipe? As we know, LLMs utilize training data from across the entire web. That means they have extensive knowledge of cookie recipes. They also have a deep understanding of how the ingredients we cook with interact with one another.

That said, if you ask for a "cakey" chocolate chip cookie, the LLM will likely rely on its recipe knowledge…

Read more

The AI 80/20 rule that you should flip

Have you ever heard of the Pareto Principle? It is a way for us humans to notice patterns. Here is how I see people use AI:

• 80% of people write task-oriented prompts

• 20% of people use AI to plan their business growth proactively

It is time to flip those numbers and turn AI into a strategic edge for your business.

Too many prompts are task-oriented

Think about your last few AI prompts. I bet they looked like this:

• Draft a blog or social post

• Clean up an email

• Brainstorm ideas for an event

• Make a presentation less boring

While valuable, those will help you complete a to-do list, not build your business.

The spinach you don’t see

Consider this: Hours after lunch, you attend meetings, chat with people in the hallway, and lead your weekly team meeting. Then, on a trip to the bathroom, you look into the mirror, smile, and staring back at you is a Texas-size spinach leaf attached to your front tooth. Everyone saw it, and no one told you!

That gut-punch? That “what else am I missing?” moment? That’s what it’s like when your competitors use AI to build their business.

Only 20% of prompts focus on long-term growth.

If 80% of people use AI to write for them, the 20% are doing something much more powerful: identifying gaps in their business, developing growth plans, and improving connections with their customers.

Since large language models are trained on nearly every part of the internet and fine-tuned to be experts in various fields, it makes sense that you can pair up with a virtual expert to extract value.

Let’s look at a few ways you can use AI growth prompts:

• Build personas based on your customers.

• Create user journeys for those personas to take them from a cold lead to a customer.

• Identify missing content on your website that may be causing lost sales.

• Build a detailed plan to grow your market share.

Those are, as they say, just the tip of the iceberg. AI provides reasoning and research models that can “think” through how to help you market, sell, and grow your business by finding your blind spots or opening new doors to revenue streams.

Don’t just prompt, reason

Before you go any further, turn up AI’s brainpower with Reasoning.

Reasoning will pull insights by “thinking” about the best way to help you improve your business.

Here is how leading AI tools handle it:

ChatGPT (OpenAI): Use any “o” model. o3 is best right now

Claude (Anthropic): Use Extended

Gemini (Google): Use any Pro model

M365 Copilot (Microsoft): Use Analyst agent

Got it, Bill. Show me.

Let’s say you’re an upstart interior designer. You’ve got a few clients, but their budgets are tight, and their requests come in at the last minute. You’re working with limited time and money to grow your business. Every day feels like building sandcastles while the tide is coming in. Can AI help you escape this cycle?

Be in the top 20% of growth prompters with these “reasoning” prompts.

I’m using an interior designer as the example here, but these prompts work no matter what business you’re in. Don’t just skim through the prompts. Repurpose them for your business today.

Diagnose your business health: Profit visibility lets you decide what to pursue and drop.

Prompt: Act as a small-business analyst. Here are my last 12 months of revenue by client, service type, and lead source. Identify which activities yield the highest profit per hour and which drain time.

Read more

If you are reading this series in its entirety, it is because you decided to go on a journey with me to install open-source AI models on your local computer.

In this final installment, I provide you with some ideas and tips for using your new open-source AI model on your computer, including:

• Create AI agents on your computer, for free

• Understanding how ope…

Read more

Welcome to Part 2 of my series, which shows you how to set up an open-source large language model (LLM) on your computer. In Part 1, I showed you how to install the open-source Ollama product that can read nearly any AI model. You also learned it has no user interface (other than a black & white terminal window). In this article, I will help you install a product called Open WebUI, and you will feel very comfortable using it if you have already used ChatGPT, Claude, Gemini, or Copilot.

I use a local AI model when working with privileged and private information, to try out new AI features, and to create or test local AI agents (a topic for another day).

Let’s go!

What you’ll see when it’s working

The one-time setup process to install Open WebUI is more work than your typical software download. First, I want to show you what it looks like, so you know it is worth your patience.

Most of the features you are familiar with from using Claude, ChatGPT, Gemini, or Copilot are all here:

• Prompt and chat in the same way you do with ChatGPT, Claude, Gemini, or Copilot

• Select a model (more on that in Part 3 of this series)

• Access your chat history

• Upload files for analysis

• Search the web

• Use voice input for conversations

• Use voice input for text-to-speech (TTS)

Simpler paths with strings attached

You will have to do some technical things to install Open WebUI. I will try to make this article as easy as possible. However, you can install at least a few dozen other products on your computer with a single setup.

So why don’t I show you those here?

Some apps have license restrictions that might concern you. Read their licensing terms carefully. Others provide interfaces designed for software developers and don’t feel like a general-use tool. Still others may work on Windows, Mac, and Linux, but the app developers focus more on one operating system over another.

If you want to skip this article and check them out, here are the ones I have been playing around with, but at this time, I am not formally recommending them:

• AnythingLLM

• Msty

• LM Studio

Mission safety first

As with most open-source software, installing it takes a little work. I promise the one-time setup is the hardest, but I will walk you through it all, and then you will be up and running!

There are two options to install Open WebUI on your computer:

1. Quick install using a technology called Docker (this article)

2. Advanced install if you have the Python programming language installed (not this article)

I will do a quick explainer so you understand the differences, but if you want to jump right into it, scroll to the “Step 1: Install Docker” section.

Before I explain Docker, let me explain the common challenges with open-source software.

For most apps you run on your computer, you select the product from an app store, download it, and run it. That can also be the case for open-source software, but not always.

Many open-source apps require you to install the underlying technology upon which they are built. For example, the Open WebUI team uses the Python programming language. Not only do you need to install that, but you will also have to install related components, called packages, that will install on your computer.

When you install Python and various packages, they get updated with new features and security updates. That means:

• You must be vigilant about keeping your computer more secure

• To “uninstall” the open-source product, you must know all the packages, programming languages, tools, and folders to remove

• If there are new features, you may have to install new versions of Python and packages, followed by the open-source product (like Open WebUI).

There must be a better way, right?

Choosing Docker to avoid the mess

Wouldn’t it be nice if you could secure all those programming languages, packages, and open-source software into an isolated container and run it? If you don’t want it anymore, you can delete the container. That is what Docker provides.

Docker is not open-source, but it is free for scenarios like yours, where we want to run open-source software without the messiness of all those products installed on our computer.

Here are the benefits you receive over the more technical approach:

• Docker containers run in isolation from your host operating system (macOS, Windows, Linux, etc.)

• No need to install all the programming and open-source tools required by Open WebUI on your computer

• Want to uninstall Open WebUI? You delete the Docker container in one easy step.

Want to learn more about Docker? Please scroll to the end of this article and watch the first 8-ish minutes of a YouTube video I created for aspiring AI software developers (I promise, the first 8 minutes will not be technical.)

Step 1: Install Docker

Read more

Learn more and sign up now

You already know the challenges of juggling brand guidelines, subject matter expertise, pressing deadlines, and the demands of managing structured and unstructured content.

You might spend hours translating complex technical details into everyday language, validating each fact, and ensuring nothing slips through the cracks. This course focuses on your specific needs by showing you how AI can streamline repetitive tasks and enhance the overall quality of your work.

Open enrollment starts now for our Spring and Summer 2025 courses

Over seven weeks, you will participate in 90-minute sessions that blend hands-on practice with strategic discussions. Each lesson will show you how to leverage generative AI and adapt it to your established style guides and domain terminology.

Learn more and sign up now

You will learn to shape prompts that produce accurate drafts, incorporate brand voice, and identify missing documentation. With thoughtful and deliberate use of AI, you can devote more of your time to tasks that require your expertise, such as improving user journeys, driving engagement, and prioritizing customer content.

• Automate repetitive tasks and accelerate content delivery using AI

• Tailor AI to your voice, audience, and content domains

• Align content with user needs and business outcomes

The Content Wrangler

Articles, interviews, event announcements, and other information to help content professionals stay up-to-date on industry trends, methods, standards, tools, and best practices.

By Scott Abel

This training program is in partnership with The Content Wrangler, so sign up for their Substack today or contact them for an exclusive discount.

I hope you join me in staying ahead of the curve with AI. -Bill

Chat with an AI on your computer, even without the Internet. For some, an open-source model can be an excellent alternative to a paid AI service. If you can prompt for something using Claude or ChatGPT, you can usually use the same prompt with a local model.

Here are some other benefits of using a local model:

• Privacy & Security: Keeps sensitive data on your device, reducing exposure to third parties

• Offline Access: Works without an internet connection

• Customization: You can fine-tune a model for personalized tasks and preferences

• Cost Efficiency: Avoids subscription fees and cloud computing costs

The AI that stays with you

Many companies and open-source groups offer lightweight AI models with capabilities similar to foundational models like ChatGPT, Claude, Gemini, and Copilot. If ChatGPT knows nearly everything about anything, a lightweight model usually knows just enough about most things.

When you download a model, it is nothing but a big file. You do not run a setup program to get it running. Instead, you need a program that can open the file and provide an interface to chat with the AI.

There are thousands of AI models. Some of the more popular models may be from companies you already know, like:

• Mistral is a family of solid models from the French company Mistral.ai

• Microsoft Phi is a family of lightweight AI models from Microsoft.

• Google Gemma is a family of lightweight AI models from Google.

Finding the right AI for you

Have you ever tried to parallel park a family-sized SUV in a bustling city? It just won’t fit. Do you want a small, efficient car to drive within a 5-mile radius for grocery shopping? You don’t need an all-terrain vehicle when a Toyota Prius would do just fine.

Your computer may be loaded with memory to spare or a lightweight, portable computer with limited capacity. That is why different model families exist.

In general, you should think of these models like this:

• Smaller = faster, less memory requirements, less knowledge to respond to your prompts

• Bigger = slower, more memory requirements, more knowledge to respond to your prompts

A model family typically starts with a small version, around 1GB. The larger the model, the more disk space it requires; some can be 10GB or more.

There are even specialized models. Some are particularly good at vision (reading images, real-time video recognition), whereas others focus on text processing and writing, like Claude. Still others can create images or even videos.

Not all open doors stay open

Just because a model says they are open source does not mean you can download and use it without legal review.

I selected the Mistral 7b model for this article because it uses the permissive Apache 2.0 license.

The popular Google Gemma, Microsoft Phi, and Meta Llama models have restrictions, so use them carefully.

That said, I am not a lawyer and probably got some of this wrong, so please refer to the licenses and consult a lawyer if you are using these models for anything other than personal or hobby use.

Check out my article here to learn more about open-source and domain-specific models.

A direct line to your AI

Let’s say I send you an Excel file and ask you to modify a formula. You can’t just open the file without Excel or a compatible app.

The same is true for local models. You download a big file but need an app to open it so you can chat with the model.

In our scenario, we will use the most popular open-source AI intermediary called Ollama. This product does two things:

1. Manage models: Ollama allows you to download, run, delete, and update models.

2. Enable prompt chats with models: You can type your prompt, and Ollama will send it to the model. Then, it will send the response back to you.

One letter apart but worlds away

No. Ollama is an open-source project that connects you to most so-called “open” large language models (LLMs). Meta (Facebook, Instagram, etc.) may or may not be involved in this project, but they are unrelated. It uses a permissive MIT license. Did I mention you should talk to a legal person, not me yet? If so, remember I am not a licensing expert.

No buttons, just a prompt

No. That is why this is a multi-part article. In this article, I set you up for success by setting up Ollama, downloading a local model, and testing it.

In part 2 of this three-part article, I will show you how to use another open-source product that will be very familiar to you because its interface resembles ChatGPT, Claude, and Gemini.

Getting there, step by step

You will be installing open-source software on your computer. I use the process below, but if you use a work computer, please talk to your IT representative first.

Read more

AI can help you with research, but can it be trusted? Let’s put its citation skills to the test. In this article, I will walk you through prompt tips to get the best possible results from AI when using it for research.

In this article, you will learn a few tips to:

• Use AI for research

• Use critical thinking to identify concepts you are not quite sure are fa…

Read more

ChatGPT hit the scene two years ago, and now AI is a trillion-dollar industry. Naturally, the rush is on to create the next best thing. Entrepreneurs, investors, and nearly every company worldwide are chasing their piece of the market. And who can blame them? Even slicing yourse…

Read more

Last week, I was on the panel for an AI Ask Us Anything event, with co-hosts Jennifer Hufnagel and Charly Jarrett. Interestingly, many people asked “How can I organize my chats in ChatGPT?” That question makes sense, because if we use AI regularly, those chats build up fast.

In …

Read more

That conversation led me to be a guest on Jeff Eyet's The BIG Strategy podcast. I share how AI saves me 30% of my time, allowing me to focus on business growth. And, of course, I managed to slip in helpful prompting tips you can start using today.

AI business automation can el…

Read more

We have all suffered the never-ending PowerPoint presentation, squinting at a sea of 10-point font, wondering if we accidentally signed up for an eye exam. Our brains check out before we even reach the second bullet point. Don’t let that happen to you, and try my “give it to me …

Read more