Loading
Services
Every capability of the platform - and whether it works on this deployment at this moment. 25 of 25 are live. Nothing on this page is a promise: each status is computed from the same stores the dashboards read when you load it.
Send a request; the router picks the model, agent, skill, tool or person that fits it best and tells you why.
88 targets in the catalogue, 82 of them answer on the platform's own providers.
Point any OpenAI client at /v1 with model "auto" and the answer comes from the model the router chose.
77 models answer here; streaming and tool calls included.
Tokens, price per candidate model and the recommended pick for a request - no account needed.
Free without a key (rate limited per address); metered as an estimate with one.
One request becomes a composition - a persona, a skill package and the model to run them - with a step-by-step trace.
Included from the payg plan.
A request that needs a person is queued for one; the answer feeds the learners like any other outcome.
1 human target in the catalogue.
Route, estimate, recommend, explain and report outcomes from Claude, Cursor or any MCP client over JSON-RPC.
Served at /mcp with the same key as the API; tool list readable without one.
Report whether an answer was good; the strategies re-weight and the next decision gets better.
Included from the free plan.
A small language model trained on routing outcomes votes in the ensemble; the autopilot retrains it and promotes a better challenger.
Trained model loaded; autopilot off.
Bring your own keys - Azure OpenAI, OpenAI, Bedrock, Vertex, Ollama or any compatible endpoint - and the router executes on them.
Included from the free plan.
Every decision has a trace you can open: the signals, the strategies' votes, the ranking, the execution and its cost.
Tracing is on; traces are scoped to your workspace.
What routing saved against always sending to the most expensive model, per request and per month.
Computed from your metered usage; the public status page shows the platform-wide figure.
Deny lists, allowed regions, monthly budgets, prompt-injection and PII guards - set per workspace, enforced on every request.
Input guard on, PII redaction on, output guard flag.
Per-tenant constraints and budgets, a hash-chained audit trail of every decision, circuit breakers per target.
Included from the team plan.
Budget, error-rate, quota and provider alerts to e-mail, Slack, Teams or a webhook, plus the in-app inbox.
Evaluated every 60 s.
A daily, weekly or monthly digest of what your workspace routed, what it cost and what it saved, in your time zone.
Delivered through m365.
Ask the assistant (it answers from the documentation through the router) or open a ticket a person answers.
932 documentation passages indexed; the assistant runs in router mode.
Agents, skills, personas, prompts and MCP tools you can install into your routing pool - or publish yourself.
224,675 published listings from 79,065 publishers.
Every routing target with measured traffic, the reference LLM catalogue with list prices and the published-benchmark leaderboard.
445 reference models, refreshed 3.0 times a day.
Daily notes on new models, LLM releases, agent frameworks and research, written through the router from the sources we follow.
Feeds polled every 3 h; posts publish automatically.
One e-mail a day (or a week) with the new posts and what the platform runs for you; registered people get it with their account.
Arrives at 09:00 in each reader's time zone; new accounts are enrolled once their address is confirmed.
Sign in with GitHub, Google, Microsoft or GitLab; organizations bring their own IdP with e-mail-domain login and just-in-time membership.
Shared providers: microsoft, google, github. Organization SSO from the enterprise plan.
Start free, upgrade in the dashboard, pay by card or from a prepaid credits wallet; every metered row is priced.
Card payments through Stripe; plans: enterprise, pro, team.
One line installs the osr CLI; the Python and TypeScript SDKs speak to this platform with the same key.
curl -LsSf https://opensmartroute.ai/install.sh | sh - then osr login --url https://api.opensmartroute.ai.
The same router as a service on your own infrastructure: Docker image, Compose bundle, Helm chart and the Azure Marketplace package.
Zero-dependency Python package; osr serve, the image and the chart ship with every release.
See what a prompt costs before you send it on ChatGPT, Claude, Gemini and the provider consoles; one shortcut opens the router anywhere.
Free and open source; they sign in with your account and count under Apps in the dashboard.
Every service above is included with a free account - create one and your first API key is ready in a minute. Registered people also receive the daily newsletter, which reports what changed here; it can be switched off at any time.
Programmatic access: GET /api/v1/services (no key required, cached for a minute). Readiness lives on the status page; the platform guide explains each field.