Questions, answered.

Oximy is one platform to see every AI tool and dollar across your company, run your team's AI work in an app you own, and optimize every request across models.

Start where the pain is clearest. Visibility maps sprawl, Sidekick gives your team a company-owned AI app, and Relay optimizes model traffic. Each product works alone, and they get stronger together.

Yes. Visibility, Sidekick, and Relay each stand on their own. Most teams start with the product that maps to their sharpest problem, then add the others as the value compounds.

Yes. Visibility is the observability layer for your AI. It keeps the map current on its own as tools, teams, accounts, workflows, and spend change.

Visibility can combine managed endpoint coverage with supported provider, identity, billing, and spend integrations. Capture, redaction, retention, and access remain under company control.

Teams can move from company-wide adoption and tool inventory into people, departments, workflows, official and shadow spend, governance activity, reviews, requests, and policies.

No application SDK or code change is required for endpoint coverage. Provider integrations can be added where deeper account or usage evidence is useful.

Identity resolution and capture controls associate company activity with the right account, device, person, and department while separating personal activity where policy and local rules require it.

Sidekick is the company-owned workspace for repeatable work, approved context, shared assets, and managed access. Specialist tools can remain where they are useful.

Yes. Admins choose the approved model menu, and teams can use the right model directly or let Relay choose the path behind a workflow.

Teams can publish prompts, agents, skills, plugins, templates, knowledge, and connectors so useful work becomes reusable company infrastructure.

Provider keys remain centrally managed. Employees use approved models without copying secrets into browsers, prompts, or side projects.

Yes. Long-running work can continue in isolated cloud sandboxes, subject to the limits and controls configured by the company.

Relay predicts which model should handle a request based on the workload, quality target, cost, latency, and routing policy you choose.

Relay routes each request to the model that clears your quality bar at the best cost and latency. Shadow mode and evals let teams prove results on real traffic before switching production requests.

Relay is focused on the routing decision. It can sit behind or alongside gateway infrastructure, but the core job is deciding which model should handle each request.

Yes. Relay can sit in front of coding agents and other high-volume workloads, routing each request to the model that meets your quality bar at the best cost and latency.

Each product is priced on its own, and you can combine them into a platform plan. Full plan details and current pricing live on the pricing page.

Relay charges for the routing layer: routed tokens and custom routers. The model calls themselves still run through your own provider accounts.

A custom router is a saved routing setup for a specific workload, model pool, objective, or environment. Plans include a base routing path, with additional custom routers available as an add-on.

No. Oximy is built to help teams use AI well, not to police them. The goal is to understand the tools, accounts, workflows, and model traffic your company runs on so better defaults can be made available.

No. Customer data is not used to train shared models, and provider retention settings can be enforced through policy.

Yes. Regional and private networking options are available to support residency and enterprise infrastructure requirements.

Yes. Relay records provider, model, policy, quality, latency, and cost signals for every routing decision.