Skip to main content

blog

Notes on AI, models and the work they do.

Comparisons, prompting technique and product updates from the LazySusan team.

40 postsnewest first
  1. 9 min readguide

    One API for Every AI Model: How to Switch Between GPT, Claude and Gemini Without Rewriting Code

    Call GPT-6, Claude and Gemini with one API key and SDK, switch models by changing one string, and add fallbacks that survive a provider outage.

  2. 8 min readguide

    How to Give Every Developer Their Own AI API Key (and See Who Spends What)

    Stop sharing one AI API key. Invite your team, give each developer and service its own key, and see spend by person, project, environment and model.

  3. 8 min readguide

    AI API Budgets: How to Set Spend Limits per Developer, Project and Environment

    Cap AI API spend per key, per developer and per team, restrict models by environment, and get alerts before a budget runs out. With code for 402s.

  4. 8 min readguide

    Image and Video Generation API: FLUX, Nano Banana, Midjourney, Veo 3.1 and More From One Key

    Generate images with Nano Banana, FLUX, GPT Image and Midjourney, and video with Veo 3.1 and Runway, from one API and SDK. Code, prices and pipelines.

  5. 8 min readguide

    AI API Cost Tracking for Engineering Managers: A Practical Guide

    Track LLM API costs by feature, team and model: keys per service, request tags, a 15-minute weekly review, unit economics and the usual cost leaks.

  6. 8 min readguide

    Which AI Model Should You Use? A Routing Cheat Sheet for Chat, Image and Video (2026)

    Which AI model to use for each task in 2026: chat, image and video cheat sheets with model ids, when to step up, and a simple router in code.

  7. 10 min readcomparison

    ChatGPT Business vs Claude Team for a Small Team: What 5 People Actually Cost

    ChatGPT Business and Claude Team both cost $25 per seat a month. A 5-person team pays $125 for one, $250 for both. Cheaper options and trade-offs.

  8. 10 min readguide

    The Cheapest Way to Use ChatGPT, Claude and Gemini in 2026 (Costed Out)

    ChatGPT Plus, Claude Pro and Google AI Pro cost $60 a month together. Five cheaper setups, costed out for light, regular and heavy use.

  9. 9 min readguide

    How to Move Your ChatGPT History to Another AI (Claude, Gemini, or Anywhere Else)

    How to export your ChatGPT chats, what's in the ZIP, and where you can import it (Gemini, Claude, LazySusan). What carries over and what doesn't.

  10. 10 min readguide

    Sora Is Shut Down. What to Use Instead for AI Video (October 2026)

    OpenAI turned off the Sora API on September 24. Here's what to use instead, which model fits which job, and what each clip costs.

  11. 10 min readcomparison

    Is Paying for ChatGPT Plus and Claude Pro Worth It? The Math for 2026

    ChatGPT Plus and Claude Pro together cost $40 a month. Here's who should keep both, who should drop one, and what a month of real usage costs instead.

  12. 13 min readcomparison

    GPT-5.5 Just Dropped. I Tested It Against Claude Opus 4.6 for a Week. Here's Who Won.

    OpenAI released GPT-5.5 with a 1M token context window and it costs more than GPT-5.2. We ran it head-to-head against Claude Opus 4.6 on coding, writing, research, and reasoning. The results weren't what we expected.

  13. 11 min readopinion

    Your ChatGPT Conversation Just Used a Bottle of Water. Nobody Wants to Talk About It.

    Every AI query costs water. A lot of it. Data centers are draining aquifers, and the companies behind your favorite AI tools are hoping you never do the math.

  14. 12 min readguide

    ASU Students Are Paying $20/mo for ChatGPT. Here's How to Get 50 AI Tools for $8.

    Arizona State University students are overpaying for AI. ChatGPT Plus costs $20/mo for one model. LazySusan's student plan gives you 50+ AI tools for $8.25/mo with your .edu email.

  15. 14 min readguide

    ASU's AI Policy Just Changed. Here's Exactly What You Can and Can't Do.

    Arizona State University updated its AI academic integrity guidelines for 2026. We broke down every rule, by college and course level, so you know exactly where the line is.

  16. 11 min readguide

    UCF Students Are Spending More on AI Than Textbooks. Here's the $8 Fix.

    University of Central Florida students are stacking AI subscriptions they can't afford. ChatGPT Plus, Claude, Midjourney — it adds up fast. There's a smarter way.

  17. 13 min readguide

    UCF's 2026 AI Policy: The Complete Guide for Students Who Don't Want to Get Expelled

    University of Central Florida updated its AI guidelines for 2026. Some professors allow it, some will fail you for it. Here's the actual breakdown by college and course type.

  18. 7 min readopinion

    I Used AI to Do My Entire Job for a Week. Here's Who Should Be Worried.

    Anthropic just mapped every job AI can replace. Programmers, customer service, data entry — the numbers are brutal. I tested it myself. Here's what actually happened.

  19. 6 min readopinion

    I Cancelled All My AI Subscriptions. Here's Why Most of Them Were a Waste of Money.

    ChatGPT Plus, Midjourney, Jasper, copy.ai — I was paying $247/month for AI tools. 95% of companies see zero ROI from AI. The entire industry has a problem nobody wants to talk about.

  20. 14 min readopinion

    Texas A&M Accused 17 Students of Using AI Last Semester. Most Were Innocent.

    AI detection tools are ruining students' academic records at Texas A&M. False positives, flawed software, and professors who don't understand the technology. Here's what happened.

  21. 10 min readguide

    Aggies Are Overpaying for AI by $900/Year. Here's the Math.

    Texas A&M students running ChatGPT Plus, Claude Pro, and Midjourney are spending $1,000+ a year on AI. The student plan math doesn't lie.

  22. 11 min readguide

    UF Students Are Burning $1,000/Year on AI Subscriptions. Gators Deserve Better Math.

    University of Florida students are paying $20/mo each for ChatGPT, Claude, and Midjourney. One subscription replaces all of them for $8.25/mo with your .edu email.

  23. 13 min readguide

    UF's AI Policy Is Stricter Than You Think. Here's What Actually Gets You Flagged.

    University of Florida's 2026 AI guidelines are clear on paper and confusing in practice. We talked to students who got caught and students who didn't. Here's the difference.

  24. 11 min readguide

    Top 10 Reasons to Use Manus AI in 2026 (And How It Stacks Up Against Perplexity Computer)

    Meta's $2B acquisition of Manus just changed the AI agent game. Here are 10 reasons Manus is the autonomous AI agent to watch in 2026 — and how it compares to Perplexity Computer, Claude Cowork, and more.

  25. 4 min readguide

    Your Typos Are Costing You Better AI Responses (And You Don't Even Notice)

    That misspelled prompt you just sent to ChatGPT? It probably gave you a worse answer. Here's how typos mess with AI outputs and what you can do about it.

  26. 5 min readnews

    OpenAI Just Raised $110 Billion. Can ChatGPT Finally Catch Claude?

    OpenAI closed the largest private funding round in history. But with ChatGPT losing enterprise market share to Claude, Pentagon controversies, and 700K cancelled subscriptions — is $110B enough?

  27. 4 min readguide

    6 Things Nano Banana 2 Does Better Than Other Image AI

    I've been testing Nano Banana 2 against Midjourney, DALL-E, and Flux. Here's where it actually wins—and where it falls short.

  28. 3 min readguide

    Make a Free Business Voicemail (Takes 10 Seconds)

    Generate a professional voicemail greeting for your business. Type your script, download the MP3. Free, no signup.

  29. 14 min readcomparison

    GPT-5.3 vs Claude Opus 4.6: I Tested Both For 2 Weeks (Plus: Why I Ditched Perplexity)

    OpenAI's GPT-5.3 and Anthropic's Claude Opus 4.6 are here. After extensive testing, here's which AI actually wins—and why Perplexity users are switching to LazySusan.

  30. 9 min readguide

    14 Ways I Actually Use ChatGPT Every Day in 2026

    Forget the generic listicles. These are the ChatGPT use cases I rely on daily—from debugging code at 2 AM to meal planning when I'm too tired to think.

  31. 10 min readguide

    16 Things Google Gemini Does Better Than Other AI

    Gemini isn't just ChatGPT with Google branding. Here are the specific use cases where it genuinely outperforms everything else.

  32. 7 min readguide

    7 Things Claude Does Better Than Every Other AI

    Claude isn't the flashiest AI, but for certain tasks it's unmatched. Here's exactly when and why I reach for Claude over ChatGPT or Gemini.

  33. 9 min readguide

    9 Practical Ways to Use Nano Banana, Google's Image Editing Model

    Nano Banana is Google's image model, and it's best at editing photos you already have. Nine practical uses, which version to pick, and what each costs.

  34. 8 min readguide

    12 Ways Perplexity AI Changed How I Research

    Search engines are dead. Here's how I use Perplexity for everything from market research to planning trips—plus why I keep backups ready.

  35. 5 min readguide

    4 Mind-Blowing Things You Can Do With Veo 3

    Google's Veo 3 is the first AI video generator that doesn't look obviously fake. Here's how I'm using it—and where it still falls short.

  36. 5 min readopinion

    Perplexity Keeps Going Down. Here's What I Switched To.

    After the third Perplexity outage this month, I finally found a better setup. It has everything Perplexity offers—plus 50 other AI models.

  37. 6 min readnews

    Claude's "You're Absolutely Right!" Problem (And Why Elon Called It Evil)

    Anthropic's Claude has a sycophancy problem that's driving developers crazy. Plus, Elon Musk just called it 'misanthropic and evil.' Here's what's actually going on.

  38. 8 min readcomparison

    I Tested GPT-5 vs GPT-4o After OpenAI's Disastrous Rollout

    OpenAI removed GPT-4o without warning and Reddit exploded. After testing both models extensively, here's what's actually better—and what got worse.

  39. 10 min readcomparison

    ChatGPT vs Claude vs Gemini: An Actually Honest Comparison

    Every comparison article picks a winner. This one doesn't. Here's what each AI is actually best at, based on months of daily use across all three.

  40. 7 min readguide

    Best AI for Coding in 2025: A Developer's Honest Take

    I've used Claude, ChatGPT, Gemini, and Copilot for real projects. Here's what actually works, what's overhyped, and what I use daily.

get started

Try the models you just read about.

Every model from OpenAI, Anthropic, Google, xAI and more in one workspace.

Starter is $8/month. Cancel anytime.