Pricing

Flat plans. You keep the savings.

We make money on the efficiency we add — never a markup on your tokens. Pick a plan, point your agent at the endpoint, and keep whatever the engine saves you.

Solo

$20/ mo

For a daily coding-agent habit.

All 70+ models, one endpoint
Transparent prompt caching
Cache-aware routing
1 seat

Request access

ProPopular

$100/ mo

For heavy, all-day agent users.

Everything in Solo
Branch + escalate routing
Higher concurrency limits
Usage analytics

Request access

Team

$300/ mo

Shared cache and budgets across a team.

Everything in Pro
Shared cache across the team
Pooled budgets + seats
SSO (soon)

Talk to us

Layer X1 is in early beta — plans go live at public launch. Request access and we'll bring you on in the next wave.

Frequently asked

Do you mark up tokens?: Never. You pay a flat plan; we make money on the efficiency the engine adds — caching, routing, and branching — not a percentage of your usage.
Which models are included?: All of them — 70+ across every major provider, plus open-source. There's no per-model surcharge; routing picks the right one for each step.
What counts as a run?: Whatever your agent does between one task and the next. Layer X1 optimizes the whole trajectory, so a single run can span many model calls.
Can I bring my own keys?: Yes — bring your own provider keys or use ours. Either way, the caching and routing layer works the same.

Get started

Put the layer under your agent.

One environment variable. Your agent doesn't change — everything underneath does.

$export ANTHROPIC_BASE_URL=https://api.layerx1.dev

Get a key