In partnership with

It's been a loud few weeks.

GPT-5.6 landed. Grok 4.5 launched. Claude shipped Sonnet 5, and Fable 5 came back.

Every one of them claims to be the best.

If you're paying twenty or thirty a month for one of these, the honest question is which one deserves it.

Short answer: there is no best model anymore. There's a best one for the work you actually do.

Now a word from today's sponsor:

200+ Claude Prompts Top Professionals Actually Use at Work

Claude can be your analyst, editor, and strategist.
But most professionals are using it to fix grammar.

These 200+ Claude prompts take it from grammar tool to your most powerful AI work assistant.

Sign up for Superhuman AI and get:

  • 200+ ready-to-use Claude prompts to get real work done in minutes — researched, tested, and used by professionals at Google, Microsoft, and NASA

  • Superhuman AI newsletter (4 min daily) so you keep learning new AI tools and skills to stay ahead in your career — the prompts are just the beginning

Okay, where they actually stand.

Claude. Top of the intelligence rankings right now. Fable 5 leads the benchmarks and Opus 4.8 sits near the top. Claude's strength is careful, structured, high-quality output, and it's the best of the lot at writing that doesn't sound like a machine.

GPT-5.6. OpenAI's new line, in three tiers. It leads on long agent tasks, the kind where the AI works for hours across your tools. If your work is "go do this whole project," it's excellent.

Grok 4.5. The value play. It ranks fourth on intelligence, but it's dramatically cheaper and more efficient. It uses a fraction of the tokens for the same job. If cost is your constraint, nothing else comes close on value.

Gemini. Buried deep in Google's tools. If your whole working life is in Docs, Sheets, and Gmail, it has an unfair home advantage.

So how do you choose?

Don't start with benchmarks. Start with your week.

Which of these is closest to your actual work?

- Mostly writing, thinking, and quality output → Claude

- Mostly long, multi-step projects across tools → GPT-5.6

- Mostly cost-sensitive volume → Grok

- Mostly living inside Google Workspace → Gemini

Then pick one. One subscription. Learn it properly.

Here's the trap nearly everyone falls into.

They subscribe to two or three and use each one shallowly. They end up behind the person who knows a single tool deeply.

The distance between the models is now smaller than the distance between a good user and a lazy one.

Want a gut check? Paste this into whichever you're paying for:

Here's the kind of work I do most weeks: [describe it honestly, the actual tasks].

Given what you're best and worst at, tell me straight: are you the right tool for this, or would I be better served elsewhere? Be specific about where you'll underperform for me.

They're surprisingly honest when you ask them that way.

I build on Claude, and I'll tell you why plainly. It's the best at the work I care about. And with Cowork it goes past answering. It finishes.

But the skill of driving one tool well is what actually pays. That's Claude Mastery.

Step by step, no tech background needed.

Quick one.

Which one are you paying for right now? Reply and tell me what you use it for most.

Talk soon,
Zephyr

Keep Reading