For vibe coders with $200–300 a month to spend on AI

Stop betting your whole AI budget on one lab.

Mix Claude, Codex, GLM, MiniMax, Cursor and more into one stack. We rank every combination of up to four coding plans by the hours you'll actually get: lived reports, not vendor banners.

Catalog dated 22 Sep 2026 · checked daily · subscriptions only, no API pricing · no vendor pays for placement

Build your stack

Set a budget and how you work. Everything re-ranks as you type, and the link in your address bar saves the whole setup.

Monthly budget
$/ month
Hours a week you code with AI iTime you're at the keyboard steering agents. They can run in parallel, but you can only make use of about 1.6× this.
h / week
Work that can wait for off-peak iOnly Z.ai (GLM) has an off-peak rate: credits cost half outside Mon–Fri 07:00–11:00 London and all weekend. Every other vendor charges the same all day, so this slider only moves Z.ai plans.50%

Your timezone iZ.ai peak: Mon–Fri 07:00–11:00 your time. Off-peak 0.5× credits after that and all weekend.
Already pay for some plans?

Pick them to see how your stack ranks, or build around it.

Fine-tune
How your hours split %
Planning
Building
Admin
Minimum model quality
Planning
Building
Admin
Terminal agent
Running plans from several labs? AI Usage Tracker shows every plan's 5-hour and weekly limits in your Windows tray. Free and open source. GitHub →

Runners-up

The share link reproduces your exact setup.

The frontier: most building hours for the money

Every stack that fits your settings. The gold line is the best you can buy at each price. Past the ceiling you can't use the extra hours, so the ranking weighs model quality and provider mix instead.

one stack frontier (typical week) frontier in a bad week your #1 (with its labs) your current stack useful ceiling · your budget

Hover any dot to see the stack. Click to pin it here.

Where your hours come from

Your #1 stack, split by plan, at your off-peak share. Striped = extra hours Z.ai's half-price off-peak window gives you.

What each budget buys

The #1 stack at each budget, with your other settings unchanged. Click a row to switch to it.

Every plan, compared

Weekly building hours on a typical week (the 50th percentile). Faded icons have a cheaper or better alternative. Click a row for the sources and haircut notes.

on the plan frontier (named) off-peak stretch: Z.ai credits cost half outside 07:00–11:00 London

What changed

Catalog updates from the daily check, with the source each one came from.

How we rank

Plain rules. Every number traces to catalog.js.

Hours are ranges, not promises

Vendor banners are ceilings. We store what people actually get: a bad week (p10), a typical week (p50) and a good week (p90), cut down from the banners toward lived reports.

Quality is a grade, not a benchmark

Each plan is rated cheap, strong or frontier for planning, building and admin. There are no leaderboard scores here. Stacks that miss the minimum you set are dropped.

Mixing labs is the point

Two plans from one vendor share one reset clock, so one bad morning hits both. Stacks spread across providers score higher, and a second model family is there when one gets stuck.

Only real off-peak discounts count

Z.ai is the only vendor that publishes a night rate (credits cost half outside Mon–Fri 07:00–11:00 London). Nobody else's hours move with the off-peak slider.

You can only steer so many agents

Useful hours are capped at 1.6× your keyboard time. Past that, a bigger plan buys nothing, so the ranking weighs quality and provider mix instead.

Nobody pays for placement

Some links are referral links. The scorer never sees them. Paused plans stay listed with a tag so a stale blog post can't sell you a dead plan.

FAQ

I have $200–300 a month. Where do I start?

Put your budget into the stack builder. At $200–300 the top stacks usually pair a frontier-grade planning plan with a high-hours building plan, plus one or two more providers so you have a second reset clock and a second model family to try. Every pick is recomputed from the catalog as you change settings.

Why not just max out one Claude subscription?

One vendor = one clock and one model family: a bad morning or a thin week hits every seat at once. Mixing pools spreads reset clocks and quality grades, and no single plan is Pareto-best at every budget. Finding the best mix is the optimizer's job.

Is GLM's off-peak really half price?

Z.ai publishes a 0.5× credit rate outside Mon–Fri 07:00–11:00 London and all weekend. It stretches hours; it does not raise concurrency or enlarge the 5h credit budget. It's the only published night rate in the catalog. Everyone else charges the same all day.

What does "haircut" mean here?

Vendor banners are ceilings. Each plan's hours are cut toward what people actually see in lived reports. For example, Claude implement hours are a haircut of official Sonnet numbers, and Fable plan hours are never added on top because they share one pool.

Can I trust plans tagged "guessed" or "single-report"?

Every plan carries a source tag: official / official-clock / community / single-report / guessed. Weakly-sourced plans are marked amber in the table and flagged as the weakest link whenever they appear in a recommended stack.

Where do the numbers come from?

Every plotted number traces to catalog.js; every rule to the README's honesty section. There's no SWE-bench and no leaderboard axis. Quality is a 1–3 role grade, never a benchmark score.

Does anyone pay you to rank them?

No vendor paid to be here. Some plan links are referral links: the site may earn credits or a commission if you subscribe, and you sometimes get a discount. The scorer never sees them. Paused plans stay listed, tagged, and never appear as buyable picks.

Why show paused plans at all?

So you don't buy a dead plan from a stale blog post. Paused plans are tagged red and sort below every buyable stack; they're hidden unless you ask for them.

Free · open source · Windows

Track every plan's limits from your tray

Once you run three subscriptions, the hard part is knowing which one still has room. AI Usage Tracker puts every plan's 5-hour and weekly quota in one tray window.

  • Claude Code, Codex, Z.ai (GLM), Grok, Gemini, OpenRouter
  • Metadata-only polling; secrets stay in Windows Credential Manager
  • No installer, no admin rights. Delete the folder to uninstall
Get it on GitHub
CLAUDE
5 hr91%38m
Week58%2d
CODEX
5 hr72%2h
Week34%4d
Z.AI
5 hr24%3h
Week12%5d