Loading
90 days of the enterprise, pro and team plans free, no card needed. See the plans
Trust
Status and uptime as measured, the service levels each plan commits to and how the platform measures them, the data-processing agreement outline, every subprocessor, the security overview and how to report a vulnerability, and the legal versions in force. Nothing here is a claim the OpenSmartRoute platform cannot show you live.
Right now
From the status page: components, incidents and the monitor's own heartbeat, refreshed every 60 seconds.
Photo: panumas nikhomkhai
Service levels
The commitments come from the live plan table, the same one the pricing page and the API publish. A workspace reads its own calendar month against them at GET /api/v1/sla and on its billing page - availability is the share of its metered requests the platform answered without a server error; support is the first response to the tickets it opened.
| Plan | Monthly availability | Support | First response | Standing |
|---|---|---|---|---|
| Free | No commitment | Community support | As capacity allows | Best effort |
| Pay as you go | No commitment | Community support | As capacity allows | Best effort |
| Pro | No commitment | E-mail support | within 2 days | Target, measured |
| Team | No commitment | Priority e-mail support | within 8 hours | Target, measured |
| Enterprise | 99.9% | Dedicated support with an SLA | within 4 hours | Commitment, with service credits in the agreement |
Measured, not promised. Availability counts every metered request of the workspace in the calendar month (UTC) and the share answered without a 5xx. Requests refused for quota, policy or an invalid key are not failures; a provider you connected yourself failing is reported as such on the trace and does not count against the platform.
Credits. On the plan sold through sales, service credits for a missed month follow the enterprise agreement; the month's figure at GET /api/v1/sla is the evidence. The other plans carry targets, not commitments, as the terms say.
Self-hosted. The same package runs in your cloud or on your own servers under Apache-2.0; availability is then yours to run, and the platform's status has no bearing on it.
Data processing
The outline of the agreement hosted customers sign; your own paper is accepted where you insist. Self-hosted editions involve no processing by us, so no agreement is needed unless you grant support access to logs - then a narrow support agreement covers it.
Ask for the signed template through sales; the privacy notice describes what is stored and for how long.
Subprocessors
Annex III of the agreement. A change is announced here and in the changelog thirty days before it takes effect.
| Subprocessor | Role | Location | Data |
|---|---|---|---|
 Microsoft Azure | Hosting: the web app, the API and the database (Container Apps) | Sweden Central (EU) | Every record the platform stores; request text only in memory while a decision is made or a request is forwarded |
 Microsoft (Microsoft 365 / Azure Communication Services) | Transactional e-mail | EU | Recipient address, subject and body of the messages the platform sends; a copy kept 30 days |
| Stripe | Payments, invoices, tax calculation | United States / EU | Billing name, address, tax id, card details (never seen by the platform), invoices |
| The model provider behind the chosen target | Executing a request, only when execution is asked for | As the target declares (region and data boundary are routing constraints) | The request text and the answer; your own provider under your own contract when you connected one |
 Google LLC (Google Analytics 4) | Aggregate statistics on the public pages, subject to consent | United States | Page views on the public site; never linked to an account or joined with usage |
Security
No certification is claimed that is not held: a SOC 2 Type I is targeted and no independent security review has been completed yet; until then this page, the security review pack and an architecture call with your security team are the review. Everything below is verifiable without talking to us, and the table says which claims are. The static statements on this page were last reviewed for release 2.1.0-preview.115; the live figures above are measured when you load the page.
16 statements a reviewer checks: 12 verifiable today, 3 in progress, 1 not claimed. Every “verifiable today” row links to evidence you can open without talking to us.
| Claim | Status | Evidence, or the gap |
|---|---|---|
| Open source under Apache-2.0 | Verifiable today | Every package on PyPI and npm, the images and the chart ship with their source under Apache-2.0 - one package for the hosted platform and both editions. |
| Published threat model and security model | Verifiable today | Trust boundaries, what is and is not protected, and the hardening steps for a self-hosted router. |
| Safety-routing red-team suite on every change | Verifiable today | Paraphrase, gadget and steering attacks against safety routing gate CI and every release (osr safety --learned-guard). |
| Automated security scanning | Verifiable today | Static analysis (bandit) over the SDK and the platform API, dependency audits (pip-audit, npm audit) and the red-team suite, on every change and weekly. |
| Hash-chained audit trail per workspace | Verifiable today | Decisions and outcomes appended to a chain you can export as CSV or JSONL; a changed or missing row is detectable. |
| Prompts and answers not stored by default | Verifiable today | The platform meters tokens, cost, target and latency and keeps the decision trace; payloads are kept only when a workspace turns on log_payloads. |
| Single sign-on (OIDC / SAML) and SCIM 2.0 provisioning | Verifiable today | Organizations connect their identity provider; Entra ID, Okta, OneLogin and JumpCloud push and remove members through SCIM 2.0 (RFC 7643 / 7644) with a provisioning token issued per organization. |
| Two-factor sign-in for people and operators | Verifiable today | Time-based codes with recovery codes for workspace users and for operators of the console; an operator can reset a locked-out person's factor. |
| Provider keys from a vault, never in the platform database | Verifiable today | A provider connection may reference an Azure Key Vault secret (keyvault://...); the API reads it with its managed identity when the client is built and the value is never written to the database. |
| Data boundary and region pinned by policy | Verifiable today | A workspace or tenant policy pins a data boundary (for example EU) and a region; a model outside it is never chosen, whatever the request asks for - the trace shows the constraint that excluded it. |
| Personal data caught before it leaves the device | Verifiable today | The browser extension, the desktop app and the IDE clients redact names, addresses, numbers and secrets on the device, before the text reaches the router or a model. |
| Public status, incidents and an availability SLA | Verifiable today | Live checks every minute, served requests per day, incidents with their timeline and an RSS feed; each paid plan states its availability commitment and the enterprise agreement carries the credit schedule. |
| Signed container images and published image scans | In progress | Images are built by a pipeline from tagged releases on pinned bases, with the static analysis and dependency audits above. Signing with provenance attestations and a published vulnerability scan per image are not shipped yet. |
| Independent security review or penetration test | In progress | The review pack - scope, evidence, reproduction steps, questions - is prepared for a third party; no independent review has been completed yet. The report is linked here when it exists. |
| SOC 2 Type I | In progress | Targeted. Until a report is issued, the trust page, the security review pack and an architecture call with your security team are the review. |
| HIPAA, ISO 27001, PCI DSS | Not claimed | No certification or attestation is held. Industry pages describe controls you configure - data boundaries, regions, PII handling, audit - not compliance you inherit. |
GET /api/v1/governance scores the controls the customer owns - keys, budgets, policy - and GET /api/v1/governance/report?format=markdown is the evidence pack an auditor asks for; the platform's own items (audit chain, guard, tracing) are reported separately as what the platform reports.Legal
Every change is announced 30 days before it takes effect; the history is public and a signed-in person sees the version they accepted.
In force since Oct 1, 2026
In force since Sep 21, 2026
Terms, privacy notice, changelog, who runs it and, for the performance side of the case, the benchmark and the live rankings - all eight proofs are indexed on one page.
For the business
A trace with every answer
Which model answered, why it won, what it cost and what the alternatives were - kept for the retention period you set.
Open a traceRedacted before it leaves
Names, numbers and identifiers are removed on your side before the request reaches a provider; prompts are not stored by default.
The security pageHash-chained audit trail
Decisions, policy changes and sign-ins land in a tamper-evident log, and the governance report summarises them for the people who ask.
What is checkable todayThen ask sales for the signed agreement and the security questionnaire pre-filled from the security review - or route a request and read the trace while you wait.
Free plan, no card. Fifteen thousand decisions a month with the full trace.
Support assistant
Answers from the documentation; checking which model is on duty
Hi - I answer from the OpenSmartRoute documentation: routing, the API, plans and quotas, self-hosting. Ask away, or open a support ticket if you need a person.