Skip to content
DecisionNodeDecisionNdedocs
  • Guides
  • API reference
  • Examples
  • Playground

start here

  • QuickstartGet startedGet an API key, send one request with three questions, and branch your code on the typed answers. Plain HTTPS, nothing to install.
  • POST /v1/decideAPI referenceAnswer typed questions about a state, optional images and video frames. One request, one buffered JSON response, one answer per question.
  • QuestionsConceptsQuestions say what to decide. Each one has a type that fixes the shape of its answer: a choice from your options, a score on your scale, a…
  • ConfidenceConceptsProbabilities are calibrated per question type, so a threshold means what it says on the data we measure.
  • ImagesConceptsSend up to 16 images and text in the same request. The model reads printed and handwritten text, amounts, dates, objects and layout, and…
  • Batch jobsPatternsSend up to 10,000 requests in one file and collect the answers later, at a lower price than live calls.
  • Pricing and billingEvery workspace runs on a plan: Starter, Pro, Business or Dedicated.
↑↓ moveopen7 suggestions
Get API keyGet API key
DecisionNodeDecisionNde

Get started

  • Introduction
  • Quickstart
  • CLI
  • Playground
  • Console and keys
  • With coding agents
  • MCP server
  • Examples

Concepts

  • State
  • Questions
  • Choice
  • Score
  • Truth
  • Number
  • Points and boxes
  • Images
  • Video
  • Boosters
  • Collections
  • Confidence
  • Determinism
  • API keys

Knowledge

  • Overview
  • Documents
  • In a decision
  • Pricing and your data
  • When a rule does not appear

Models

  • DecisionNode
  • DecisionNode Flash
  • DecisionNode Edge
  • Limits
  • Versions
  • Dedicated capacity

Fine-tuning

  • Overview
  • Prepare your dataset
  • Upload and validation
  • Start a training run
  • Watch a run
  • The quality gate
  • Use your model
  • Limits and pricing

Patterns

  • Confidence-gated routing
  • Form router
  • Fan-out
  • Guardrails
  • Control loops
  • Batch jobs
  • Keys on devices
  • Bill your customers per device
  • Publishable keys

Integrations

  • Home Assistant
  • MQTT bridge
  • ROS 2

API reference

  • Overview
  • POST/v1/decide
  • POST/v1/systemone
  • GET/v1/models
  • POST/v1/uploads
  • /v1/keys
  • GET/v1/keys/{key_id}/usage
  • GET/v1/usage
  • POST/v1/credit
  • Errors
  • Safety check
  • Rate limits

Sessions API

  • Sessions overview
  • POSTOpen a session
  • WSStream frames
  • DELEnd a session

Collections API

  • Overview
  • POSTCreate a collection
  • GETList collections
  • GETRetrieve a collection
  • DELDelete a collection
  • POSTEnroll a face
  • GETList items
  • PATCHUpdate an item
  • DELDelete an item
  • POSTAdd a document
  • PUTReplace a document
  • GETRetrieve an item
  • GETList passages
  • DELDelete failed documents
  • POSTSearch a knowledge base

Batch API

  • The batch object
  • POSTCreate a batch
  • POSTAdd requests
  • POSTFinalize a batch
  • GETRetrieve a batch
  • GETGet batch results
  • POSTCancel a batch
  • GETList batches

Pricing and billing

  • Pricing and billing
  • Refer & earn

Policies

  • Responsible use
  • Data and privacy
  • Benchmarks
  • Pricing
  • Playground
Get API key
  • Guides
  • API reference
  • Examples
  • Playground

Get started

  • Introduction
  • Quickstart
  • CLI
  • Playground
  • Console and keys
  • With coding agents
  • MCP server
  • Examples

Concepts

  • State
  • Questions
  • Choice
  • Score
  • Truth
  • Number
  • Points and boxes
  • Images
  • Video
  • Boosters
  • Collections
  • Confidence
  • Determinism
  • API keys

Knowledge

  • Overview
  • Documents
  • In a decision
  • Pricing and your data
  • When a rule does not appear

Models

  • DecisionNode
  • DecisionNode Flash
  • DecisionNode Edge
  • Limits
  • Versions
  • Dedicated capacity

Fine-tuning

  • Overview
  • Prepare your dataset
  • Upload and validation
  • Start a training run
  • Watch a run
  • The quality gate
  • Use your model
  • Limits and pricing

Patterns

  • Confidence-gated routing
  • Form router
  • Fan-out
  • Guardrails
  • Control loops
  • Batch jobs
  • Keys on devices
  • Bill your customers per device
  • Publishable keys

Integrations

  • Home Assistant
  • MQTT bridge
  • ROS 2

API reference

  • Overview
  • POST/v1/decide
  • POST/v1/systemone
  • GET/v1/models
  • POST/v1/uploads
  • /v1/keys
  • GET/v1/keys/{key_id}/usage
  • GET/v1/usage
  • POST/v1/credit
  • Errors
  • Safety check
  • Rate limits

Sessions API

  • Sessions overview
  • POSTOpen a session
  • WSStream frames
  • DELEnd a session

Collections API

  • Overview
  • POSTCreate a collection
  • GETList collections
  • GETRetrieve a collection
  • DELDelete a collection
  • POSTEnroll a face
  • GETList items
  • PATCHUpdate an item
  • DELDelete an item
  • POSTAdd a document
  • PUTReplace a document
  • GETRetrieve an item
  • GETList passages
  • DELDelete failed documents
  • POSTSearch a knowledge base

Batch API

  • The batch object
  • POSTCreate a batch
  • POSTAdd requests
  • POSTFinalize a batch
  • GETRetrieve a batch
  • GETGet batch results
  • POSTCancel a batch
  • GETList batches

Pricing and billing

  • Pricing and billing
  • Refer & earn

Policies

  • Responsible use
  • Data and privacy
  1. docs
  2. /
  3. Get started

Console, keys and members

The console is where a workspace keeps its API keys, members, balance and usage. How keys are created, rotated and revoked, what each role may do, and what deleting a workspace means.

on this page10 sections
  1. Workspaces
  2. Members and roles
  3. API keys
  4. Rotating a key
  5. Usage
  6. Boosters and fine-tuning
  7. Start
  8. Plan
  9. Billing
  10. Deleting a workspace
Workspace
Its members share its API keys, its prepaid balance and its usage
Keys
Start with dn_live_, are shown once, and work with both models and every endpoint
Roles
Owner, Admin, Developer and Billing
Test keys
None: every key is live and billed (see Plans). Try requests in the playground
Where
Console: Home, Playground, Usage, API keys, Billing and Settings

Workspaces#

A workspace is the unit of access and billing. Everyone in it shares its API keys, its prepaid balance and its usage charts, and every request any of its keys sends is drawn from the one balance. Rate limits are the workspace's too, shared by all its keys; a key may carry its own lower limits, so one device or service cannot use up the rest (see Rate limits). Rename a workspace under Settings; the name, 2 to 40 characters, shows in the sidebar and on receipts. Under Settings an Owner or Admin can also turn off DecisionNode Flash fallback (the fallback policy never), so a DecisionNode request is never answered by DecisionNode Flash when DecisionNode has no ready capacity, but gets a 529 to retry instead (see No ready capacity).

Members and roles#

  • Owner

    May
    Everything, including deleting the workspace
  • Admin

    May
    Manage keys, members, billing, settings and boosters
  • Developer

    May
    Create and revoke keys, read usage, fine-tune and deploy models
  • Billing

    May
    Manage the balance, auto reload and invoices
What each role may do
RoleMay
OwnerEverything, including deleting the workspace
AdminManage keys, members, billing, settings and boosters
DeveloperCreate and revoke keys, read usage, fine-tune and deploy models
BillingManage the balance, auto reload and invoices
  • Invite people by email under Settings, with the role they need. Give developers the Developer role and the finance team Billing.
  • Change a role at any time from the members list.
  • Removing a member ends their access to the workspace at once. Keys they created keep working until you revoke them, so revoke a person's keys when they leave, or rotate them as below.

API keys#

  • Create a key under API keys, choose What this key can do (its kind: Full access, Decisions only, Manage device keys or Publishable), and give it a name of up to 48 characters, such as the service that will use it. The full secret is shown once, when you create it. Store it in your secret manager or in the DECISIONNODE_API_KEY environment variable straight away: afterwards the console keeps only its last four characters.
  • A key reaches what its kind allows. A Full access key works with both models, /v1/decide, sessions, batch jobs, collections, /v1/models and per-key usage; a Decisions-only key only decides and lists the models, and searches collections only with May search your face collections ticked; a Publishable key only decides. Those two may carry their own rate limits and a daily spend cap, always on for a Publishable key. A Manage device keys key only makes, changes and revokes Decisions-only keys and reads keys' usage. The kind is set when the key is made. Batches and sessions belong to the key that created them.
  • Track use. The keys table shows who created each key, when, and when it was last used. Search by name, id or last four.
  • Revoke a key from its row. Revoking is immediate and cannot be undone: the next call with it gets 401, and a session stream it connects closes with code 4401.
  • Rate limits of your workspace are on the Rate limits tab.

No test-mode keys

Every key is a live dn_live_ key and every answered request is billed. To try request shapes, use the playground: it needs an account and no key. A real request is cheap: 62 input tokens on DecisionNode Flash cost a fraction of a cent.

Rotating a key#

Rotate a key when someone who held it leaves, when it may have leaked, or on your own schedule. Two keys can be live at once, so rotation needs no downtime:

  1. 1.Create a new key, named for the service and the date.
  2. 2.Deploy it: put it in your secret manager and roll your service so every instance reads it.
  3. 3.Check the switch: the old key's last used stops moving, and Usage filtered by key shows traffic only on the new one.
  4. 4.Let what belongs to the old key finish: its open sessions, and the results of its batch jobs, which only the key that created them can read.
  5. 5.Revoke the old key.

If a key leaked

Revoke it first and rotate after. A revoked key stops at once, and requests made with it before then are billed to the workspace.

Usage#

Usage charts spend, input tokens and requests for the workspace, and says how it pays: from your prepaid balance, or, on a dedicated workspace, billed on your monthly invoice. Filter by model (or all models), by key (revoked keys included), and by range: 24 hours, 7 days or 30 days, read hourly (up to 7 days) or daily (7 days or more). Spend is input tokens at each model's price, or your agreed rates, and booster calls: what your workspace is charged, the same as your invoice and per-key usage (at agreed rates, a couple of minutes behind). Output is free. In code, usage.input_tokens on each response is exactly what that request billed.

Boosters and fine-tuning#

Boosters lists the boosters your requests may name, what each reads and costs, and a switch for each. Every booster is on for a new workspace; an Owner or Admin turns one off or on, and the change applies within a second. Fine-tuning holds your fine-tuned models: their datasets, training runs, scorecards and deployments. Every member can see both pages; Owners, Admins and Developers can train, deploy and cancel a run. To label past requests, an Owner or Admin turns on Keep requests for labelling: answered requests are then kept for 30 days in the workspace's storage location (see Label your history).

Start#

Start is where a workspace without a plan begins: after sign-up, an Owner, Admin or Billing member chooses how it starts before the rest of the console opens. Starter and Business start at once, monthly or yearly, without a trial. A Developer of a workspace with no plan sees that the workspace has no plan, and asks its owner. A workspace whose trial was cancelled or whose plan ended comes back to this step. See Plans.

Plan#

Plan shows the workspace's plan. During the trial: the days left, the usage of the trial's cap, and the date and amount it converts at, with Start (unavailable) now, Switch to Starter and Cancel trial. On a plan: the allowance used this period and the day it resets, today's booster test calls and what the plan includes. Every member can see it; an Owner, Admin or Billing member upgrades at once, downgrades or cancels at the period end, and undoes a waiting change with Keep my plan. Where a plan limits what you are doing, a booster outside it, a knowledge base or a batch job below Pro, fine-tuning below Business, a key or member past the count, the console says which plan includes it and what it costs each time you try. A Developer gets Ask your workspace owner, which sends the request to the Owners, Admins and Billing members, on their Plan page and in a daily email, with a message to copy as well. See Plans.

Billing#

Billing holds the balance, top-ups, auto reload and one receipt per payment. What a request costs and how the balance works is on Pricing and billing.

Deleting a workspace#

Only the Owner can delete a workspace, under Settings, by typing its name to confirm. Every key stops working at once, the remaining balance is forfeited, and it cannot be undone. Download what you need first: receipts, usage, and the results of any batch jobs.

Creating a key means you accept the Terms of Service and the Acceptable Use Policy on behalf of your organisation.

previousPlaygroundnextWith coding agents

DecisionNode is built and run by Bynn Intelligence, Inc.

  • Home
  • Playground
  • Examples
  • Console
  • Responsible use
  • Terms
  • Acceptable use
  • Privacy
  • Data processing
  • Defence addendum
  • Cookies
  • Affiliate program

on this page

  1. Workspaces
  2. Members and roles
  3. API keys
  4. Rotating a key
  5. Usage
  6. Boosters and fine-tuning
  7. Start
  8. Plan
  9. Billing
  10. Deleting a workspace