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. Models

Limits

Context size, image and booster limits, question limits, body limits, session and batch limits, and the default rate limits. These apply to both models, and to fine-tuned models built on them.

on this page3 sections
  1. Sessions
  2. Batch jobs
  3. Throughput
  • Context per request

    Value
    64k tokens (65,536)
    Notes
    State, images and every question together
    Over the limit
    400 max_tokens_exceeded
  • Images per request

    Value
    Up to 16
    Notes
    JPEG, PNG or WebP, base64 in data or an https url. Within the token limit: an image is about one token per 32 x 32 pixels, so sixteen at full size do not fit
    Over the limit
    400 api_usage_error
  • Image size

    Value
    Up to 10 MB each
    Notes
    Measured on the image bytes; base64 text is about a third longer
    Over the limit
    400 api_usage_error
  • Image pixels

    Value
    Up to 40 megapixels each
    Notes
    Resized so the longest side fits 2,048 px; metadata stripped. See Images
    Over the limit
    400 api_usage_error
  • Options per choice

    Value
    1 to 255
    Notes
    Two or more recommended
    Over the limit
    400 Too many choices
  • Levels per score

    Value
    1 to 10
    Notes
    Lowest first; two or more recommended
    Over the limit
    400 Too many score levels
  • Videos per request

    Value
    Up to 4, sent as frames with timestamps
    Notes
    Up to 128 frames over all the request's videos. The server reads frames at a small size by default (about 480 x 256 for 16:9). See Video
    Over the limit
    400 naming the limit
  • Boosters per request

    Value
    Up to 4, each named once
    Notes
    age, ai_generated, face_detection, face_recognition, ocr, bynn_depth, bynn_segment and knowledge; a time budget of 2 s for age, 5 s for ai_generated, 2 s for face_detection, 2 s for face_recognition, 12 s for ocr, 800 ms for bynn_depth, 1.5 s for bynn_segment and 800 ms for knowledge; run side by side before the model. See Boosters
    Over the limit
    400 api_usage_error
  • Values per number question

    Value
    Up to 256 as one question, 4,096 in all
    Notes
    Any step, such as 0.5 or 0.1, with at most 6 decimals in min and step; a finer or wider grid is composed in the same call. A min, max or step that is not a number (float_type), min above max or a step of 0 or below is also a 422, naming the field
    Over the limit
    422 too_long
  • Questions per request

    Value
    No count limit
    Notes
    The token limit is the only bound: 300 short questions about a short state are answered in one call. Every question, option and level takes at least one token
    Over the limit
    400 max_tokens_exceeded
  • Question keys and option names

    Value
    No length limit of their own
    Notes
    Any string, the empty string included; returned byte for byte, untrimmed and with case kept. Keys are bounded by the body size, option names by the token limit
    Over the limit
    Never refused for length alone
  • Request body

    Value
    Up to 25 MiB
    Notes
    Images sent inline as base64 count toward it; images by upload_id or url do not, so send large images as uploads. A body without images is bounded by the token limit long before this: one too large to fit gets 400 max_tokens_exceeded before it is parsed
    Over the limit
    413, answered before the body is read
  • Nesting

    Value
    At most 128 objects and arrays inside each other
    Notes
    The body's own object counts; so every value can be echoed back, such as a level in legend
    Over the limit
    400 api_usage_error
  • Body arrival

    Value
    A byte at least every 20 s, the whole body within 300 s
    Notes
    Upload time does not count toward answering
    Over the limit
    408 request_timeout_error
  • Response

    One buffered JSON body On /v1/decide every answer arrives together; sessions stream one reply per frame

Request limits
LimitValueNotesOver the limit
Context per request64k tokens (65,536)State, images and every question together400 max_tokens_exceeded
Images per requestUp to 16JPEG, PNG or WebP, base64 in data or an https url. Within the token limit: an image is about one token per 32 x 32 pixels, so sixteen at full size do not fit400 api_usage_error
Image sizeUp to 10 MB eachMeasured on the image bytes; base64 text is about a third longer400 api_usage_error
Image pixelsUp to 40 megapixels eachResized so the longest side fits 2,048 px; metadata stripped. See Images400 api_usage_error
Options per choice1 to 255Two or more recommended400 Too many choices
Levels per score1 to 10Lowest first; two or more recommended400 Too many score levels
Videos per requestUp to 4, sent as frames with timestampsUp to 128 frames over all the request's videos. The server reads frames at a small size by default (about 480 x 256 for 16:9). See Video400 naming the limit
Boosters per requestUp to 4, each named onceage, ai_generated, face_detection, face_recognition, ocr, bynn_depth, bynn_segment and knowledge; a time budget of 2 s for age, 5 s for ai_generated, 2 s for face_detection, 2 s for face_recognition, 12 s for ocr, 800 ms for bynn_depth, 1.5 s for bynn_segment and 800 ms for knowledge; run side by side before the model. See Boosters400 api_usage_error
Values per number questionUp to 256 as one question, 4,096 in allAny step, such as 0.5 or 0.1, with at most 6 decimals in min and step; a finer or wider grid is composed in the same call. A min, max or step that is not a number (float_type), min above max or a step of 0 or below is also a 422, naming the field422 too_long
Questions per requestNo count limitThe token limit is the only bound: 300 short questions about a short state are answered in one call. Every question, option and level takes at least one token400 max_tokens_exceeded
Question keys and option namesNo length limit of their ownAny string, the empty string included; returned byte for byte, untrimmed and with case kept. Keys are bounded by the body size, option names by the token limitNever refused for length alone
Request bodyUp to 25 MiBImages sent inline as base64 count toward it; images by upload_id or url do not, so send large images as uploads. A body without images is bounded by the token limit long before this: one too large to fit gets 400 max_tokens_exceeded before it is parsed413, answered before the body is read
NestingAt most 128 objects and arrays inside each otherThe body's own object counts; so every value can be echoed back, such as a level in legend400 api_usage_error
Body arrivalA byte at least every 20 s, the whole body within 300 sUpload time does not count toward answering408 request_timeout_error
ResponseOne buffered JSON bodyOn /v1/decide every answer arrives together; sessions stream one reply per frame

A request over a limit is refused before any model work and costs nothing. The state is never cut to make it fit, and the body names the limit you crossed. Bodies and fixes are on Errors.

Sessions#

Sessions stream decisions over one open connection, for control loops that decide more than 10 times a second. Their limits apply on top of the request limits above.

  • Frame

    Value
    Text or JSON of at most the session's max_frame_tokens (never more than 4,096 tokens), or one image (with an optional short note)
    Notes
    The open response gives each session's value. An image frame is held to the image limits above
  • Fixed part

    Value
    Small enough to leave room for the window and the longest question; over it, 400 max_tokens_exceeded
    Notes
    The instructions, state and questions, sent once, with room left for the window
  • Window

    Value
    1 to 64 frames, 8 by default, the frame being answered included
    Notes
    Set with window when the session opens; the recent frames together hold at most window_tokens, and the oldest leave first
  • Lifetime

    Value
    300 seconds by default, 1 to 3,600
    Notes
    Set with ttl_seconds
  • Idle

    Value
    Closed after 120 seconds with no frame on an open socket
    Notes
    Close code 4408; the server's pings keep a quiet socket open until then
  • Open sessions per key

    Value
    4, unless we set yours otherwise; past it, 429 session_limit_exceeded
    Notes
    One socket per session
  • Token rate

    Value
    Frames count against your key's input-token rate, not its request rate
    Notes
    Over it, a frame gets a rate_limit_error frame error and the socket stays open
Session limits
LimitValueNotes
FrameText or JSON of at most the session's max_frame_tokens (never more than 4,096 tokens), or one image (with an optional short note)The open response gives each session's value. An image frame is held to the image limits above
Fixed partSmall enough to leave room for the window and the longest question; over it, 400 max_tokens_exceededThe instructions, state and questions, sent once, with room left for the window
Window1 to 64 frames, 8 by default, the frame being answered includedSet with window when the session opens; the recent frames together hold at most window_tokens, and the oldest leave first
Lifetime300 seconds by default, 1 to 3,600Set with ttl_seconds
IdleClosed after 120 seconds with no frame on an open socketClose code 4408; the server's pings keep a quiet socket open until then
Open sessions per key4, unless we set yours otherwise; past it, 429 session_limit_exceededOne socket per session
Token rateFrames count against your key's input-token rate, not its request rateOver it, a frame gets a rate_limit_error frame error and the socket stays open

Batch jobs#

Batch jobs run a file of requests at 30% off the live price on DecisionNode and 50% off on DecisionNode Flash, with a multi-hour delay. Every request keeps the request limits above; the batch adds its own. /v1/models publishes them in each model's batch object.

  • Requests per batch

    Value
    10,000
    Over the limit
    400 api_usage_error
  • Upload per batch

    Value
    256 MiB, images included; 10 MiB per add call
    Over the limit
    413 request_too_large_error
  • One line

    Value
    25 MiB as sent, and in effect 17,767,680 bytes escaped as a JSON string, an inline image's base64 included
    Over the limit
    413 request_too_large_error; nothing from the call is applied
  • Active batches per key

    Value
    10: draft, queued or running
    Over the limit
    429 batch_limit_exceeded
  • Requests in active batches per key

    Value
    100,000 across draft, queued and running batches
    Over the limit
    429 batch_limit_exceeded
  • Unfinished requests

    Value
    Expire 24 hours after the batch is finalized
    Over the limit
    Not billed; resubmit them
  • Results

    Value
    Kept 7 days after the batch ends
    Over the limit
    410 results_expired
Batch limits
LimitValueOver the limit
Requests per batch10,000400 api_usage_error
Upload per batch256 MiB, images included; 10 MiB per add call413 request_too_large_error
One line25 MiB as sent, and in effect 17,767,680 bytes escaped as a JSON string, an inline image's base64 included413 request_too_large_error; nothing from the call is applied
Active batches per key10: draft, queued or running429 batch_limit_exceeded
Requests in active batches per key100,000 across draft, queued and running batches429 batch_limit_exceeded
Unfinished requestsExpire 24 hours after the batch is finalizedNot billed; resubmit them
ResultsKept 7 days after the batch ends410 results_expired

Throughput#

Rate limits are your workspace's, shared by all its keys and both models: by default 40 requests and 100,000 input tokens a second, and a key may carry its own lower limits. Your workspace's values are in the console. Over a limit, the API answers 429 with Retry-After and retry-after-ms. See Rate limits.

Need more?

Dedicated capacity runs on GPUs reserved for you, with limits set to your traffic. See Dedicated capacity for what changes and how to ask for it.

previousDecisionNode EdgenextVersions and retirement

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. Sessions
  2. Batch jobs
  3. Throughput