open source · self-hosted · zero dependencies

The command center for every AI project you run.

A dependency-free LLM gateway and self-hosted dashboard. Point any project at it - any language, one command - and get cost, traces, prompt versions, budgets and anomaly alerts across your whole AI portfolio. No SDK to adopt, no database to run.

Get startedLive demo →$ npx ai-command-center
<1ms proxy overhead0 runtime deps11+ providersMIT licensed
Interactive sample dashboard. Toggle display currency and time range below.
localhost:4321 - dashboard
Total spend
₹2,812.39
Requests
4,669
Tokens
22.6M
Latency p50
3.6s
Spend over time
30 Juntoday
Spend by project
claims-copilot₹1,975.73
invoice-extraction₹457.92
catalog-enrichment₹193.66
support-chatbot₹185.08
01 / Integration

Change one base URL. Keep your keys. Ship.

Every provider SDK already supports a custom base URL, so onboarding a project is one line - or zero, via an environment variable. Nothing else about your code changes, and your API keys are forwarded to the provider untouched.

Python
from openai import OpenAI

# the only change: point base_url at the gateway
client = OpenAI(base_url="http://localhost:4321/p/invoice-bot/openai/v1")
# your OPENAI_API_KEY is passed straight through, untouched
HOW IT WORKS

Your app calls the LLM as before, but through the gateway. It forwards the request untouched, streams the response straight back, and reads token usage on the side to compute cost. Added latency is well under a millisecond - a thin, auditable proxy, nothing more.

OpenAIAnthropicGeminiMistralDeepSeekxAIGroqTogetherOpenRouterOllama+ any OpenAI-compatible

Also: install a thin helper (pip install aicc-sdk or npm i @ai-command-center/sdk) that sets the base URL for you; group calls without touching the path via an x-aicc-project header; report batch or unsupported-provider usage to /api/track; and swap /p/<project> for a /k/<gateway-key> when auth is on. Full integration guide →

02 / Capabilities

More than a cost meter.
A control room.

Every capability runs from the same one-command install, with metadata-only storage and no external services.

COST & USAGE

Spend across your whole portfolio, in your currency

Per-request cost from real token counts (including cached tokens), grouped by project and model, with latency percentiles and error rates. Shown in ₹ / $ / € with live exchange rates and stored in USD.

₹ / $ / €live FX, INR-first
p50 / p95latency per model
cachedtokens priced right
liveSSE request feed
Traces

Session timelines

Group the calls in one request or agent run with a trace header, then read the timeline.

Prompts

Version tracking

Tag prompts by version and compare cost, latency and error rate as they change.

Budgets

Limits & alerts

Per-project monthly budgets, error-rate and latency thresholds, and an optional webhook.

Anomalies

Spike detection

Explainable rules flag cost spikes and error bursts per project - no ML, no black box.

Models

Head-to-head

Effective cost per million tokens, p50/p95 latency and error rate, compared side by side.

Routing

Failover & balancing

Opt-in routes fail over and load-balance across providers - no client change, no new dependency.

Privacy

Keys & data stay put

Provider keys pass through and are never logged. Prompt and response bodies are never stored.

03 / Positioning

Lightweight on purpose.

Platforms like Langfuse, Helicone and LangSmith go deeper on tracing and evaluation; LiteLLM and Portkey are richer gateways. They are also a database, a queue and an analytics cluster to operate. This is one command and a file of JSONL - the answer to “what is each project spending, and is anything off?” without standing up infrastructure.

Reach for a full platform when you need distributed span trees, LLM-as-judge evals or advanced routing policies. Reach for this when you want honest cost and usage visibility - plus opt-in failover routing - across many projects, self-hosted, in minutes. See the fact-checked comparison →
04 / Roadmap

Shipped, and what's next.

Clear about what exists today and what is deliberately not built yet.

shippedCost & usage dashboardSpend, tokens, latency and errors per project, in INR, USD or EUR.
shippedTraces / sessionsGroup an app request or agent run into a call timeline.
shippedPrompt versioningTrack templates by version; compare cost, latency and error rate.
shippedBudgets & alertsPer-project monthly budgets, threshold alerts, optional webhook.
shippedAnomaly detectionRule-based cost-spike and error-burst flags, per project.
shippedModel comparisonEffective cost/1M tokens, p50/p95 latency and error rate.
shippedProvider routingOpt-in failover and load-balancing across providers.
shippedRoles & project grantsRead-only viewer role and per-user project access.
nextSSO (OIDC / SAML)Single sign-on beyond the built-in accounts.
nextManaged optionA hosted deployment for teams that would rather not self-host.

Give every AI project a command center.

No signup, no database, no vendor. One command and a base URL.