---
title: "Pricing for reliable AI agent orchestration runtime"
description: "Pricing for procurement on Duale AI's agent orchestration runtime, with Evaluation, Production and Enterprise tiers, AIOp metric, and model inference billed by your providers."
lang: en
lastUpdated: 2026-06-07
url: https://duale.ai/en/product/pricing
---

## AI-generated summary

Provides Duale AI’s early‑access pricing for its runtime that orchestrates AI agents, defines the AIOp metric for billing runtime work, and details the three tiered plans (Evaluation, Production, Enterprise) along with the commercial review process that finalizes pricing, deployment boundaries, and provider setup.

- The pricing separates runtime work (orchestration, retries, state transitions, policy checks, review records) from model inference, which you pay directly to your providers.
- Evaluation tier provides early‑access to the Python SDK, a managed workspace, basic task identifiers and event signals, and a scoped AIOp volume; pricing is available on request.
- Production tier is usage‑based, adds durable orchestration, recovery, and requestable review inputs; model inference is billed directly by your providers.
- Enterprise tier provides custom pricing, deployment‑boundary and retention reviews, multi‑project/team support, and security & privacy review aligned with rollout scope.
- Commercial review sets the AIOp volume, counting rules, support level, deployment boundary, retention, and provider setup, and creates the order form that governs runtime billing.

Summaries were generated by AI. Generative AI is experimental.

---

<Hero
  eyebrow="For procurement and finance"
  title={
    <>
      Pricing for <AccentText>reliable agent orchestration</AccentText>
    </>
  }
  subtitle="Pricing separates Duale runtime work from model inference: orchestration, retries, state transitions, policy checks, and review records where configured. Model inference is billed by the providers you choose."
  primaryCta={{ label: "Discuss your rollout", dialog: "contact" }}
  secondaryCta={{ label: "Security posture", href: "https://duale.ai/en/product/security.md" }}
/>

<Section
  title="Early-access plans"
  subtitle="Plans below are indicative for design partners and early-access customers. Final commercial terms depend on deployment boundary, retention, support, and provider setup."
>
  <PricingTable>
    <PricingTier
      title="Evaluation"
      tagline="For the first Python agents"
      price="Contact us"
      features={[
        "Early-access Python SDK",
        "Evaluation workspace on managed runtime",
        "Task identifiers and basic event signals where configured",
        "Proposed AIOp volume scoped during review",
        "Community and product-team feedback channel",
      ]}
      primaryCta={{ label: "Request early access", dialog: "contact" }}
    />
    <PricingTier
      title="Production"
      tagline="For teams moving agents out of pilots"
      price="Usage-based"
      period="/ AIOp volume, scoped during review"
      features={[
        "Durable orchestration and recovery paths",
        "Requestable review inputs for production reviews",
        "Configurable provider routing policy",
        "Model inference billed by your providers",
        "Security and deployment review before scale-up",
      ]}
      highlight
      highlightLabel="Recommended"
      primaryCta={{ label: "Discuss your rollout", dialog: "contact" }}
    />
    <PricingTier
      title="Enterprise"
      tagline="For regulated and multi-team portfolios"
      price="Custom"
      features={[
        "Deployment boundary and retention review",
        "Support for multiple projects, teams, and providers",
        "Security, privacy, and procurement review package",
        "Commercial terms aligned with rollout scope",
      ]}
      primaryCta={{ label: "Talk to sales", dialog: "contact" }}
    />
  </PricingTable>
</Section>

<Band>
  <Section
    title="What the proposed AIOp unit measures"
    subtitle="AIOp is a proposed commercial unit for runtime work. The order form must define the final metric and included volume."
  >
    <CardGrid columns={3}>
      <Card title="Included" icon="check">
        Agent orchestration, retries, state transitions, event emission, policy checks, and review records where
        configured.
      </Card>
      <Card title="Excluded" icon="layers">
        Model inference, vector databases, cloud infrastructure outside the managed service, and third-party tools you
        contract directly.
      </Card>
      <Card title="Why it matters" icon="bar-chart">
        The useful cost is not the cheapest model call. It is the cost of a reliable result that can be operated,
        reviewed, and improved over time.
      </Card>
    </CardGrid>
  </Section>
</Band>

<Section
  title="Cost model"
  subtitle="Model pricing and capabilities change quickly. The production cost moves to routing, reliability, supervision, and lifecycle control."
>
  <CardGrid columns={3}>
    <Card title="Bring your providers" icon="id-card">
      Keep commercial control of model providers and switch them as capabilities, prices, and compliance requirements
      change.
    </Card>
    <Card title="Route by policy" icon="mixer-horizontal">
      Decide when a small model is enough, when a stronger model is justified, and when another automated attempt should
      stop for project review.
    </Card>
    <Card title="Scale a portfolio" icon="rocket">
      Move from a few pilots to many production agents without building separate visibility, retries, decision gates,
      and review inputs for each one.
    </Card>
  </CardGrid>
</Section>

<Section title="Buying notes" subtitle="The price conversation should map to the production boundary.">
  <CardGrid columns={2}>
    <Card title="Good fit" icon="check">
      Teams that need durable Python agents, several model providers, task identifiers, available operating signals,
      review inputs, and a shared operating model between engineering, security, and business teams.
    </Card>
    <Card title="Current boundary" icon="exclamation-triangle">
      Teams looking for a finished no-code agent builder, a marketplace of prebuilt assistants, or a turnkey packaged
      deployment without design-partner work.
    </Card>
  </CardGrid>
</Section>

## Pricing questions

### What does the AIOp unit cover and what does it exclude

AIOp meters the Duale AI runtime work that makes an agent operable: orchestration, retries, state transitions,
    policy checks, runtime event emission, recovery handling, and the records the runtime attaches to a run for review.
    It excludes model inference, vector databases, customer cloud infrastructure, and third-party tools you contract
    directly with their providers.

### Who pays for model inference

Unless separately agreed in the order form, you choose and contract with your model providers, and they bill you
  directly for inference. Duale AI pricing covers the managed runtime and does not bundle a provider fee. External
  databases, customer cloud, and third-party tools are priced separately.

### Is AIOp the final billing metric

Yes. AIOp is the billing metric Duale AI uses for runtime work. The order form fixes the counting rules and included
  volume for your deployment alongside retention, support, deployment boundary, and provider setup.

### Is the Production package the right starting point

Pick by stage. Evaluation is for a first Python agent on the evaluation workspace with a scoped AIOp volume.
  Production is for teams moving an agent out of pilot into operated workflows, with runtime usage, routing policy, and
  review inputs. Enterprise is for regulated rollouts or multi-team portfolios that need a deployment boundary,
  retention, and procurement review. Commercial review picks the package against your workflow scope and decision gate.

### What does commercial review decide

Commercial review fixes the AIOp volume and counting rules, the support level, the deployment boundary, the
    retention duration, the provider setup, and any rollout-specific legal and commercial terms. The output is the
    order form.

<Cta
  title="Price the runtime before the model bill surprises you."
  primaryCta={{ label: "Discuss your rollout", dialog: "contact" }}
  secondaryCta={{ label: "Review security posture", href: "https://duale.ai/en/product/security.md" }}
/>

<JsonLd
  data={{
    "@context": "https://schema.org",
    "@type": "OfferCatalog",
    name: "Duale AI pricing",
    description:
      "Early-access pricing for durable AI agent orchestration, runtime events where configured, recovery, policy checks, and requestable review inputs. Concrete price, proposed AIOp metric, and included-volume terms are scoped during commercial review.",
  }}
/>

## Related content

- [Production runtime for durable AI agents](https://duale.ai/en/home.md)
- [Turn AI pilots into a production portfolio](https://duale.ai/en/solutions/business.md)
- [Build Python agents with routing, retries, and task IDs](https://duale.ai/en/solutions/developers.md)
- [Role-based solutions for platform, business, and governance](https://duale.ai/en/solutions.md)
- [Govern AI agents in production with shared operating signals](https://duale.ai/en/solutions/governance.md)
- [Terms and conditions for SaaS service](https://duale.ai/en/legal/cgv.md)

---

## Sitemap

See the full [Markdown sitemap](https://duale.ai/sitemap.md) for all pages.