Skip to content

Enterprise

Shared application memory across hundreds of engineers.

One org-scoped memory layer for every team and every AI coding agent.

  • SSO/SAML
  • SCIM
  • RBAC
  • Audit logging
  • Data residency
  • On-premises
  • 99.9% SLA

Ostra models

Models built for testing, governed like the rest of TestRelic.

Ostra Pro and Ostra Decision, from one key and one balance.

Live

Ostra Pro

An advanced model for testing software.

Explore Ostra Pro
Beta

Ostra Decision

Fast verdicts for CI and agents.

Explore Ostra Decision

Ostra Code

A model for writing and maintaining test code.

Explore Ostra Code

Ostra Visual

Visual regression, video included.

Explore Ostra Visual

Governed in the Console

One login, your roles

Your TestRelic login, with owner, admin, developer, billing and viewer roles.

Keys with limits

Full access or read only, an expiry, a spend cap and an IP allowlist.

An audit trail

Key, billing and member changes are logged. Request logs can be switched off.

Included, and unlimited

Ostra usage comes with your plan, and Enterprise has no usage limit.

Ostra can also run inside your own environment, by arrangement: air gapped, in your cloud or managed.

The platform

Every product, one memory layer.

From the SDK that captures a run to the agent that reads it back.

Memory

CORE

Technology

Memory of every decision your team makes about its tests.

A durable, per-repository knowledge base your AI agent writes and the whole team reads. Quarantine calls, assertion conventions, maintenance TODOs, "always / never" constraints — captured once from your Ask AI conversations and respected on every future turn.

4

memory categories

Auto

saved + categorized

Team

shared, per repo

Decision

Agreed verdicts — quarantine this spec, keep that assertion, delete the dead test. The calls your team has already made, so they aren't re-debated.

Insight

Observations about the suite — what flakes, why a flow is slow, what a failure really meant. Hard-won knowledge that usually lives in one engineer's head.

Maintenance

Follow-up TODOs and refactoring notes — what to revisit, what to clean up, what's deferred. Durable, not lost in a chat scrollback.

Context

Constraints and conventions — the "always / never / know X before touching Y" rules. So neither a teammate nor an AI agent re-breaks them.

platform.testrelic.ai/repos/shop-e2e/memory

The journey of one fact

selector.checkout.pay

[data-test=pay-now] → [data-test=pay-submit]

100%

Fresh

re-confirmed by run #4821, today

CapturedDistilledServeDecayForget
Capturedconfidence 100%1 / 5

Every run emits raw signal — traces, timings, selectors, failures — from Playwright, Appium, Maestro and pytest.

Without a memory layer: run output scrolls past · CI log expires in 30 days

Proof

Teams already shipping on shared memory.

Against the industry

Debug Time
  • TestRelic AI
  • Industry average
Weeks against percent
Model12345678
TestRelic AI9294919596939794
Industry1920222631342925

Industry: DORA 2024, four published anchors spread over eight points. Source

Triage Accuracy
  • TestRelic AI
  • Industry average
Weeks against percent
Model12345678
TestRelic AI9190939295949695
Industry6872576482504434

Industry: World Quality Report 2024, headline figures in release order. Source

Coverage Growth
  • TestRelic AI
  • Industry average
Weeks against percent
Model12345678
TestRelic AI9391949296979598
Industry7571626161605655

Industry: DORA 2024, Figure 3. Source

TestRelic AI values are illustrative.

Model benchmarks

Ostra Pro: cost to fix failing tests
  • Ostra Pro
  • Other coding models
Failing tests fixed against total cost
Model020406080100
Ostra Pro$0.00$0.82$1.64$2.46$3.28$4.09
Claude Sonnet 5.5$0.00$2.40$4.80$7.20$9.60$12.00
Claude Opus 5.5$0.00$4.80$9.60$14.40$19.20$24.00
GPT-5.6 Terra$0.00$2.48$4.96$7.44$9.92$12.40
Grok 4.7$0.00$2.24$4.48$6.72$8.96$11.20

One fix: 50,000 input and 2,000 output tokens. The lines are the total cost of 0 to 100 fixes. Other models at published list prices, read 2026-11-19.

Ostra Decision: cost to triage failures
  • Ostra Decision
  • Coding models asked the same question
Failures triaged against total cost
Model02004006008001,000
Ostra Decision$0.00$0.0001$0.0002$0.0003$0.0004$0.0005
Ostra Pro$0.00$1.55$3.10$4.64$6.19$7.74
Claude Sonnet 5.5$0.00$4.60$9.20$13.80$18.40$23.00
Claude Opus 5.5$0.00$9.20$18.40$27.60$36.80$46.00
GPT-5.6 Terra$0.00$4.72$9.44$14.16$18.88$23.60
Grok 4.7$0.00$4.36$8.72$13.08$17.44$21.80

0 to 1,000 failures. Other models read 10,000 tokens of context each; Ostra Decision is priced per task. Published list prices, read 2026-11-19.

Governance

Powerful, yet governable

Roll out to hundreds of engineers without giving up control.

Identity and access

SSO/SAML, SCIM 2.0 provisioning and role-based access down to the resource.

Read the platform docs
acme · Members14 users

SSO enforced · SAML 2.0

Oktaconnected
Entra IDAzure AD
WorkspaceGoogle

Directory · SCIM 2.0 synced

SSarahsarah@acme.comAdmin
DDanadana@acme.comMaintainer
AAmiramir@acme.comViewer
JIT provisioningon
Deprovisioningautomatic
Custom roles · resource-level permissions4 roles

Security and audit

AES-256 at rest, TLS 1.3 in transit and an exportable audit trail. You choose the region.

Air-gapped and on-prem
acme · Securityaudit · retention 1 year
GDPR AES-256 TLS 1.3

Audit log · tamper-proof

10:42:07member.role.changed@dana → Maintainer
10:41:55token.createdtr_mcp_•••• · read-only
10:38:12export.siem1,204 events

Streamed to

SplunkSIEM export
Datadogmetrics
SentinelAzure
Data residencyUS · EU · APAC
Pen testing · bug bounty · DDoS protectionactive
Vulnerability scanning · incident response24/7

Support and SLAs

A dedicated CSM, a one-hour response guarantee and a contractual 99.9% uptime.

Forward-deployed engineering
acme · Support24/7/365

All systems operational

Uptime guarantee99.9%
Support response< 1 hr
Critical resolution< 4 hrs
Planned maintenance< 4 hrs/q

Escalation path

Shared channelyour workspace
PagerDutycritical
Email · phone24/7
Dedicated CSMassigned
Escalation to engineeringdirect
Onboarding · best practices · roadmap inputincluded
Quarterly business reviewsscheduled

Deployment options

Run it wherever your policy says it has to run

Same platform, three boundaries. Choose where your data sits and who runs it.

Managed Hosting

We run the whole platform. Uptime, upgrades and scale are ours.

How managed hosting works
Managed by TestRelicacme · us-east

Operational · 41 days without an incident

Uptime99.9%
Capacityauto

Operations

Provisionedsame week
Upgradesapplied for youv3.1.2
Backupspoint-in-time recoverycontinuous
On-call rotaTestRelic SRE
Patch windowapplied off-peak

Bring Your Own Cloud

The full platform in your own cloud account. Your data, keys and IAM stay yours.

How BYOC works
Your cloud accountacme-prod

Deployed into your account · 3 regions available

AWSdeployed
Google Cloudsupported
Azuresupported

Under your controls

Runs inyour VPC, your regioneu-west-1
Encryptionyour KMS keyscustomer-managed
Accessyour IAM, your policiesunchanged
Operated by TestRelicfully managed
Counts toward committed cloud spendyes

Air Gapped (On-Prem)

Entirely on your premises, with no outside connection. The full platform, not a cut-down build.

How air-gapped works
On-premises · isolatedacme · your DC

Egress blocked at the boundary

Network egressnone
Deliverysigned offline bundleverified
Modelsinference inside your wallslocal
Capabilityfull platform, not a lite build100%

Sovereignty

Data leaving the buildingnone, ever
Update cadenceyou choose the window
Telemetry sent homenone

Forward Deployed Engineering

Our engineers, in your sprints

TestRelic engineers embed with your team, own the maintenance and deliver inside your sprints.

In your scrum, not on a call sheet

Our engineers work inside your systems, standups and access controls, so there is no hand-off.

Sprint 24 · shop-e2eyour board

Nadia joined 14 days ago · standup 9:15 daily

Your Jirasame sprint
Your Slack#eng-quality
Your reposyour reviews

On the team

N(Nadia (TestRelic)Forward Deployed EngineerFDE
SSarahEngineering LeadOwner
Accessyour SSO, your controlsscoped
Knowledge-transfer overheadnone
Delivering fromweek one

Maintenance is owned, not ticketed

Flaky suites, coverage gaps and healing are owned in-sprint while your engineers stay on features.

Owned by your FDEshop-e2e

4 closed this sprint · 0 handed back

checkout.spec.tsselector healed, re-run greenmerged
Flaky suitefailed 4 of the last 30 runs→ 0
Coverage gapsclosed in-sprint3 closed

Meanwhile

Your engineersshipping featuresunblocked
Backlog handed back to younone
Handover or extend — your callat any point

Platform-backed, not just hands

Every FDE uses the same memory, Studio, CLI and MCP tools you do. The tests and memory stay yours.

How the engagement works
TestRelic harnessnot a staffing contract

Same tools your team gets — nothing bespoke

Memoryshared
Studio
CLI
MCP39 tools

Engagement

Scoping callweek 0
Embed on-site / in-systemsweek 1
Deliver in your sprintsongoing
What stays when it endsthe tests, the memory

Ready to get started?