All spotlights

Product spotlight · June 2026

Connect OpenAI SDKs to Quant and cap AI spend for each interface

The AI platform now has OpenAI-compatible endpoints for existing SDKs, daily and monthly spend caps for each AI interface, per-user budget overrides and workflow drafts.

Updated AI platform For government and enterprise

What changed in June 2026

The June 2026 changelog has the full details.

  • OpenAI-compatible endpoints

    New /oai/v1 endpoints cover chat completions (buffered and streaming), embeddings and model listing. Quant identifies the organisation from the bearer token, so you can set the base URL of an OpenAI SDK to Quant with no other changes. Read OpenAI compatibility.

  • Spend caps for each interface

    Organisation admins can now set total daily and monthly spend caps for each AI interface in AI governance. An AI interface is the channel that requests come through, such as Slack, the API gateway, streaming, workflows, orchestrations or embeddings.

  • Per-user budget overrides

    Admins can override the per-user budget for individual users, with an Unlimited option that exempts a user from it. Spend from autonomous QuantCode jobs is now tracked under its own identity, system:code-agent, and you can give it a separate budget with an override. Read AI Governance: Spend Limits.

  • Workflow drafts and publishing

    The workflow editor now keeps your edits in a draft, so you can change a workflow without affecting live runs. Published workflows open read-only, with a prompt to edit them as a draft. For tools that start an asynchronous job, a new Await result option makes the workflow wait for the job to finish, so later steps receive its result. New and updated workflows now register GitHub webhooks correctly when you save them.

  • AI usage API

    The /ai/usage endpoint now returns the same AI usage figures as the subscription page, grouped by model, user or token, with an optional 30-day daily series. It needs a token with the new subscription:read scope, and the token owner must have permission to manage the organisation's subscription. This is a breaking change: the endpoint needed AI access and the ai:use scope before, and its response format has changed. /ai/usage/me is unchanged.

  • Terraform and SDKs

    The quant_ai_governance Terraform resource supports per-interface spend limits and per-user overrides. Version 4.19.0 of the TypeScript and Go clients covers the new spend-limit fields. getAIUsageStats moves to the AI API and returns the new /ai/usage response, and the scaling policy methods also change.

By the numbers

endpoints for OpenAI SDKs
/oai/v1
of daily AI usage from the API
30 days
TypeScript and Go clients
v4.19.0

Action needed

  • Save again any existing workflow that still returns a "webhook not configured" error.
  • Update integrations that call /ai/usage to use a token with the subscription:read scope and to read the new response format.
  • Update code that calls getAIUsageStats or the scaling policy methods when you upgrade to v4.19.0.

Who it is for

  • Application developers

    Point an existing OpenAI SDK at Quant with a new base URL, and keep your chat, streaming and embeddings code.

  • Organisation admins

    Set daily and monthly caps for each AI interface, and give individual users their own budget or an Unlimited one.

  • Workflow builders

    Change a workflow as a draft while live runs continue, then publish the draft when you are ready to use it.

How it works

The /oai/v1 endpoints let you use an OpenAI SDK with the Quant AI platform. In January 2026, the chat inference and streaming endpoints started to accept the OpenAI request format. Now the platform also has dedicated /oai/v1 endpoints. Quant identifies your organisation from the bearer token, so you can set the base URL to Quant with no other changes.

On QuantGov, set base_url to https://dash.quantgov.cloud/oai/v1.

AI governance now has these spend controls:

  • Per-user daily and monthly budgets, which you could already set.
  • Total daily and monthly caps for each AI interface.
  • Overrides of the per-user budget for individual users, with an Unlimited option that exempts a user from it.

Spend from autonomous QuantCode jobs is now tracked under the system:code-agent identity. Their spend no longer counts towards the budget of the user who owns the job’s token. To find user IDs for overrides, look at the AI usage breakdown, which now shows each user’s ID. The override field also suggests known system identities.

Get started

  1. Set the base URL

    Set your OpenAI SDK base URL to https://dashboard.quantcdn.io/oai/v1.

  2. Use the QuantGov URL

    On QuantGov, use https://dash.quantgov.cloud/oai/v1 instead.

  3. Open AI governance

    In the dashboard, open AI Services > AI Governance.

  4. Cap each interface

    Set a daily and a monthly spend cap for each AI interface that you use.

  5. Override user budgets

    Give a user their own budget, or system:code-agent a separate one.

Related: Explore AI platform · Documentation · June 2026 changelog

Ready to try it on Quant?

Use the AI platform on your own project, or talk to our team first.