Hey everyone,

Anthropic moved Claude Fable 5 to extra credit usage this week. Base plan credits still run the same — but Fable 5 now costs extra on top of that.

Everyone's complaining. Here's why you should be taking notes instead.

Anthropic moved Claude Fable 5 to extra credit usage this week. That means: you burn through your base plan credits, then Fable 5 keeps running at additional cost.

The reaction from dev Twitter: outrage. Another paywall. AI getting more expensive. The little guy locked out of the good stuff.

Here's the part everyone's missing.

What most devs do: default to the most powerful model for everything. Fable 5 for a two-line regex. Fable 5 to summarize a README. Fable 5 because it feels safer.

What senior engineers do: route each task to the cheapest model that's good enough — and prove it with evals.

That gap is real money. At any company running AI at scale, the difference between "always use the top model" and "route intelligently" is hundreds of thousands of dollars in annual API spend. That's a line item someone's always tracking.

Your move: Spend one week routing deliberately. Haiku for fast lookups and short summaries. Sonnet for standard code review and analysis. Fable 5 only when the task genuinely needs deep multi-step reasoning. Track what actually needed it. Spoiler: less than you think.

There’s a second way AI projects leak money that nobody mentions. It’s not the API bill but it’s the billing layer you never built. Most devs spend a week wrestling Stripe and still ship something half-broken. Tabs fixes it: usage-based billing for AI apps, wired up in an afternoon. You’re an intrigued developer who should absolutely take a look at it.

Slow billing doesn't just create friction — it delays cash and compounds across every deal you close.Slow billing doesn't just create friction — it delays cash and compounds across every deal you close.

Most SaaS finance teams assume the gap between contract signature and first invoice is a minor inconvenience. The Tabs Billing Lag Calculator puts a dollar figure on it.

Answer 5 quick questions about your contracts, invoicing process, error rate, and DSO, and the calculator benchmarks your billing lag against top SaaS companies — then shows you exactly what it's costing you.

Two minutes. One number that's hard to ignore.

Calculate your billing lag and see where you stand.

See your number

In a technical screen, most candidates say they'd "use the best available model." The candidate who says "I'd profile the task distribution, route to the cheapest sufficient model with evals to validate quality, and reserve Fable 5 for complex reasoning steps" is describing what senior engineers actually do.

One sentence. Completely different signal.

Anthropic moving Fable 5 behind extra credits is them being honest: running their best model is expensive, and someone has to pay for it. That pricing signal is the same cost pressure you'll navigate in every AI engineering role.

The students who understand that NOW — before they graduate — walk into their first job already thinking like a senior engineer.

The paywall isn't the obstacle. The paywall is the lesson.

Until next time,
– The Jobless Team