I’ve mentioned before that I am NOT handy.

About 30 years ago (yikes!) when I first married, I tried to impress my wife by “installing” closet shelves for my wife.

I spent two hours at Home Depot buying a cartful of pieces and parts for shelving and hangers.

That was Saturday morning.

Saturday night during dinner, we heard a loud crash in our bedroom.

The entire closet had imploded on itself.

My father-in-law came by two days later with a steel pipe.

In one minute, he wedged it between two studs, and the closet was fixed.

I learned a valuable lesson.

Solutions don’t need to be fancy and expensive—they should be as simple as they need to be.

All the pay-as-you-go LLM models have a large “pro” model and a smaller low-priced model.

So let’s do a quick comparison between Gemini and OpenAI.

We want to create a social media post every day.

If every social media post involves back-and-forth conversation, we can assume that each exchange includes additional tokens for both input and output.

Let’s estimate that each post requires 5 exchanges of 500 tokens (input) and 500 tokens (output) each, leading to a total of 10,000 tokens per post (500 input + 500 output × 5 exchanges).

Now, let’s recalculate the total yearly cost for 365 posts, with each post requiring 10,000 tokens.

So we’re assuming . . .

• Each social media post requires 10,000 tokens for back-and-forth conversation (5 exchanges).

• 365 posts per year = 3,650,000 tokens (input + output combined).

Pricing Breakdown for Each Model:

1. GPT-4o

• Input: $2.50 per million tokens

• Output: $1.25 per million tokens

• Total tokens per post = 10,000 tokens (5 input + 5 output)

2. GPT-4o Mini

• Input: $0.150 per million tokens

• Output: $0.60 per million tokens

• Total tokens per post = 10,000 tokens (5 input + 5 output)

3. Gemini 1.5 Flash

• Input: $0.075 per million tokens

• Output: $0.30 per million tokens

• Total tokens per post = 10,000 tokens (5 input + 5 output)

4. Gemini Pro

• Input: $1.25 per million tokens

• Output: $5.00 per million tokens

• Total tokens per post = 10,000 tokens (5 input + 5 output)

Cost Calculation for 365 Posts (10,000 tokens per post):

Summary of Yearly Costs (for 365 posts with back-and-forth conversation):

Conclusion:

• Gemini 1.5 Flash remains the most cost-effective option at $1.37 for 365 posts involving back-and-forth conversation.

• GPT-4o Mini costs $2.74, which is more expensive than Gemini Flash but still affordable.

• GPT-4o increases significantly in cost to $13.69 for 365 posts.

• Gemini Pro is the most expensive, costing $22.81 for the same 365 posts.

Now this is JUST for social media posts . . . not accounting for all the other stuff we do.

And when we start talking about AI Assistants and AI automation, costs can really start to add up.

So . . . should you use GPT-4o Mini or GPT-4o? Only YOU can answer that question.

You may sincerely like the content of GPT-4o WAY better.

But if a steel rod (4o Mini) works for you, then why spend extra money on fancy shelving (GPT-4o)?

Troy's One Cool Thing of the Day

Key Features:

• 🤖 AI Engine: Embedded AI handles routine project management tasks automatically.

• 🐛 Bug Triage: Automates the process of sorting and prioritizing software bugs.

• 📊 Backlog Grooming: Helps maintain and organize project backlogs without manual effort.

• 📝 Documentation Updates: Automatically keeps project documentation up-to-date.

• 👥 Team Collaboration: Facilitates better teamwork through AI-assisted project management.

The Bottom Line: Height aims to enhance project collaboration and productivity by automating routine tasks, potentially allowing teams to focus more on strategic work and less on administrative chores.

Disclaimer: Just a heads-up! Some of the links in this newsletter might be affiliate links, which means I could earn a small commission if you make a purchase. No extra cost to you… just a way to keep the magic coming!