Live — a real gateway, not a mockup

Every AI dollar your engineers spend —
visible, governed, proven.

lean-ctx Enterprise is the self-hosted control plane for AI coding usage: per-person cost attribution, an org-wide model catalog, on-the-wire context compression and savings evidence your CFO can verify. Everything below is served by a running instance — open the console and check the numbers yourself.

Why engineering orgs deploy it

One gateway between your agents and every model provider — because three problems show up in every org the month after AI tools roll out:

$“Nobody can tell me what AI costs us.”

Provider invoices arrive as one line item. The gateway meters every request per person, team, project and model into your own Postgres — the console and a printable value report answer the CFO question with measured data, not estimates.

“Every laptop talks to a different model.”

Clients call org aliases like demo/fast; which real model answers is a gateway decision. Swap the model for 800 seats with one config line — plus budgets, rate limits and a model ceiling under signed org policy.

“Prove the savings — don't claim them.”

Prompts are compressed on the wire; every event books actual cost and the counterfactual cost at a frozen reference model. Ed25519-signed evidence exports let finance and auditors verify the delta offline. On this instance, right now: measured live above.

Don't take the page's word for it — it's running

Public, read-only access to the same three surfaces a customer deployment has. The credentials below are published on purpose: the admin API is GET-only and inference endpoints are not routed to the internet.

Gateway Console org view

Spend, verified savings, the avoided-cost baseline and a live breakdown by person × project × model × provider — what a platform lead sees.

Admin token (read-only API)
86f63be6a9491ddcc372ef4adef86c2ba96e5bfc7c9caf2c1d307dc173ce90eb
Open the console →

Personal view /me — per developer

Each engineer signs in with their own gateway key and sees exactly their usage and savings — never anyone else's. Try the shared guest identity.

Guest gateway key
gk-guest-demo-leanctx-com-aefc73310ee74382dcfb9a1411fc01969243b27627589f49
Open /me →

Value report prove

The printable ROI artifact finance gets: actual spend, verified savings and the counterfactual baseline — regenerated every 15 minutes from the live ledger with lean-ctx gateway report.

Open the report →
Young numbers, honest numbers: this instance went live on July 6, 2026 and accumulates real history day by day — nothing is backfilled, so the 30/90-day charts fill up as the demo ages. It runs small open-source models priced at a transparent local shadow rate, so absolute dollar amounts are small; the savings share and the mechanics are exactly what a production deployment shows at enterprise scale.

The 30-day proof — on your traffic, at zero software cost

The gateway ships inside the Apache-2.0 open-source binary (gateway-server feature, since v3.9.0). Running the evaluation costs you nothing but a docker compose up — you find out what you'd save before you spend anything.

Deploy in a day

lean-ctx gateway init scaffolds config, secrets, keys and compose; Helm and Terraform templates exist for k8s and cloud. Runs entirely on your infrastructure — provider keys never reach laptops.

Point 5–10 engineers at it

Two env vars per IDE or agent (ANTHROPIC_BASE_URL + personal key). No plugin, no client rollout — it speaks the providers' own wire protocols.

Let it meter for 30 days

Real traffic accrues in your Postgres: per-person costs, compression savings, the counterfactual baseline against the model you'd otherwise pay list price for.

Decide on evidence

gateway report renders your value report; gateway evidence exports Ed25519-signed aggregates your finance team verifies offline. Your numbers — not our marketing.

How it works

One gateway between your coding agents and every model provider. Four verbs, all live in this demo:

Coding agents
  • Claude Code / Cursor
  • CLI agents & scripts
  • (here: a traffic seeder
    doing real coding turns)
lean-ctx gateway (this demo)
  • SEE — per-person keys, metering to Postgres
  • ROUTE — org aliases demo/fast, demo/standard
  • REMEMBER — prompts compressed on the wire
  • PROVE — savings vs. a frozen reference model
Model providers
  • here: self-hosted Ollama
    (open-source models)
  • in production: Anthropic,
    Azure AI Foundry, OpenAI, …
How local models are priced: self-hosted inference is never booked at $0 — hardware and power are real. Local usage is costed at a transparent, configurable shadow rate (default $0.25 per million tokens, stated in the report's methodology), so “savings” from moving traffic to local models stay honest instead of infinite.

Built for the questions security and finance ask first

Self-hosted, your data planeGateway, Postgres and models run on your infrastructure. Air-gapped deployment is supported — nothing phones home.
Provider keys off laptopsOrg keys live in the gateway's environment; engineers hold personal, individually revocable gateway keys (SHA-256 hashes at rest).
Read-only admin APIThe console's API is GET-only. Budgets, policies and keys change via CLI on the host — not from a browser.
Governance under signed policyModel ceiling, per-person daily budgets and rate limits enforce with honest wire-shape errors (403/429 + Retry-After).
GDPR tooling includedPer-person export and erasure (gateway gdpr export|delete), optional pseudonymization, bounded retention.
Fail-open by designIf the metering database is briefly down, traffic keeps flowing — availability is never traded for bookkeeping.

Run this yourself in two minutes

Everything on this page is in the open-source binary — self-hosting the gateway is free.

# scaffold a complete instance (config, secrets, compose, keys)
lean-ctx gateway init ~/gateway --org="Your Org" --seats=100 --person=you@example.com

# bring it up (gateway + Postgres) and check it end-to-end
cd ~/gateway && docker compose up -d
lean-ctx gateway doctor

# point any IDE or agent at it
export ANTHROPIC_BASE_URL=http://your-host:8484
export ANTHROPIC_AUTH_TOKEN=<your personal gateway key>