Skip to content
DecisionNodeDecisionNde
  • Model
  • Inference
  • Benchmarks
  • Examples
  • Docs
  • Pricing
Sign inGet API keyGet API key
  • dnModelTyped answers, calibrated confidence
  • msInferenceOur own stack and GPUs, answers in ms
  • %BenchmarksAccuracy per suite, with intervals
  • exExamplesBuilds you can start from today
  • /v1DocsQuickstart, API reference, recipes
  • $PricingPay per input token, output is free
  • inSign inEmail link or Google, no password
benchmarks

Measured. Wins and losses.

DecisionNode against TypeSafe Jev, Cloudflare Clef and five open decision models: 15 text suites, 25,134 items, every suite shown, including the ones we lose. Then two image suites, places in photos, video and drone views, and our own latency.

Open playgroundOpen playgroundHow we measured↓
DecisionNode, 15-suite mean
0.817
DecisionNode Flash
0.783
DecisionNode Flash, short request
5 ms

Preliminary results, October 2026

The 15-suite mean

Accuracy averaged over the 15 text suites, one dot per system. DecisionNode scores 0.817 and TypeSafe Jev 0.803. DecisionNode Flash, the tier built for speed and price, scores 0.783.

  • our models
  • other systems

Preliminary results, October 2026

0.600.650.700.750.800.85
  1. 1DecisionNodeBynn Intelligenceours0.817
  2. 2TypeSafe Jev 1.13TypeSafe AI0.803
  3. 3Cloudflare Clef 27BCloudflare0.786
  4. 4DecisionNode FlashBynn Intelligenceours0.783
  5. 5Cloudflare Clef-flash 9BCloudflare0.759
  6. 6H2O-Lightning-4Bopen model0.748
  7. 7pplx-decider-27bopen model0.746
  8. 8Cygnet (Gemma 4 12B)open model0.717
  9. 9Kev-4Bopen model0.717
  10. 10Strands Decider 2Bopen model0.622

accuracy, 15 text suites

Where we lead, where we trail

DecisionNode minus TypeSafe Jev on every text suite. A bar is solid where the two 95% intervals do not overlap and hollow where the difference is inside the noise.

  • DecisionNode ahead, intervals apart
  • TypeSafe Jev ahead, intervals apart
  • inside the noise
−20−100+10+20
  1. Prompt injection0.768 vs 0.600+16.8 pts, DecisionNode 0.768, TypeSafe Jev 0.600, intervals apart
  2. Web tasks0.760 vs 0.698+6.2 pts, DecisionNode 0.760, TypeSafe Jev 0.698, inside the noise
  3. Rules and policies0.883 vs 0.831+5.2 pts, DecisionNode 0.883, TypeSafe Jev 0.831, intervals apart
  4. Public decision set0.906 vs 0.855+5.1 pts, DecisionNode 0.906, TypeSafe Jev 0.855, inside the noise
  5. Large option sets0.813 vs 0.775+3.8 pts, DecisionNode 0.813, TypeSafe Jev 0.775, inside the noise
  6. Familiar task families0.793 vs 0.774+1.9 pts, DecisionNode 0.793, TypeSafe Jev 0.774, inside the noise
  7. Multilingual0.801 vs 0.782+1.9 pts, DecisionNode 0.801, TypeSafe Jev 0.782, inside the noise
  8. Typed decisions0.740 vs 0.735+0.5 pts, DecisionNode 0.740, TypeSafe Jev 0.735, inside the noise
  9. Short classifications0.899 vs 0.904−0.5 pts, DecisionNode 0.899, TypeSafe Jev 0.904, inside the noise
  10. Long inputs0.846 vs 0.853−0.7 pts, DecisionNode 0.846, TypeSafe Jev 0.853, inside the noise
  11. Transfer tasks0.844 vs 0.852−0.8 pts, DecisionNode 0.844, TypeSafe Jev 0.852, inside the noise
  12. Open decision tasks0.788 vs 0.798−1.0 pts, DecisionNode 0.788, TypeSafe Jev 0.798, inside the noise
  13. Held-out decisions0.932 vs 0.948−1.6 pts, DecisionNode 0.932, TypeSafe Jev 0.948, inside the noise
  14. New task families0.788 vs 0.823−3.5 pts, DecisionNode 0.788, TypeSafe Jev 0.823, intervals apart
  15. Knowledge (MMLU-Pro)0.692 vs 0.820−12.8 pts, DecisionNode 0.692, TypeSafe Jev 0.820, intervals apart

← TypeSafe Jev aheadDecisionNode ahead →

difference in accuracy

Leads

Clear leads on prompt injection (+16.8 pts) and rules and policies (+5.2 pts). Ahead too on web tasks (+6.2 pts) and public decision set (+5.1 pts), by margins inside the noise. That is the everyday work of a workflow: hold to the task when the input argues, apply a written policy and act on a page.

Trails

Jev is clearly ahead on knowledge (−12.8 pts) and new task families (−3.5 pts), and ahead by a margin inside the noise on held-out decisions (−1.6 pts) and open decision tasks (−1.0 pts). If your decisions lean on broad world knowledge, weigh that suite first.

Every suite, every system

Accuracy per suite, with the 95% interval under it for our models and TypeSafe Jev. The highest score on each row is marked. Scroll sideways on a small screen.

Accuracy by suite and system, with 95% intervals where published
SuiteDecisionNodeBynn IntelligenceDecisionNode FlashBynn IntelligenceTypeSafe Jev 1.13TypeSafe AICloudflare Clef 27BCloudflareH2O-⁠Lightning-⁠4Bopen modelpplx-⁠decider-⁠27bopen modelCygnet (Gemma 4 12B)open modelKev-⁠4Bopen model
Held-out decisions636 items0.9320.91 to 0.950.9360.92 to 0.950.9480.93 to 0.960.9330.9260.959 (highest on this suite)0.9240.892
Public decision set117 items0.906 (highest on this suite)0.84 to 0.950.8120.73 to 0.880.8550.78 to 0.920.8550.8720.8500.8680.744
Transfer tasks730 items0.8440.81 to 0.880.7790.74 to 0.820.852 (highest on this suite)0.81 to 0.890.8100.7530.8290.7860.762
Web tasks1,106 items0.760 (highest on this suite)0.73 to 0.790.7510.72 to 0.780.6980.66 to 0.730.6970.6120.6720.6320.612
Knowledge (MMLU-Pro)1,479 items0.6920.67 to 0.710.5720.55 to 0.600.820 (highest on this suite)0.80 to 0.840.6340.4820.6270.5440.494
Short classifications178 items0.8990.84 to 0.940.8650.81 to 0.920.9040.86 to 0.940.9210.8310.941 (highest on this suite)0.9130.792
Open decision tasks5,011 items0.7880.78 to 0.800.7700.76 to 0.780.7980.79 to 0.810.834 (highest on this suite)0.8090.7900.7590.811
Typed decisions955 items0.740 (highest on this suite)0.71 to 0.780.7130.68 to 0.750.7350.70 to 0.770.7050.6880.7200.6630.663
New task families3,629 items0.7880.78 to 0.800.7500.74 to 0.760.823 (highest on this suite)0.81 to 0.830.8000.7760.8000.7580.738
Familiar task families4,100 items0.793 (highest on this suite)0.78 to 0.810.7670.76 to 0.780.7740.76 to 0.790.7720.7280.7770.7270.707
Large option sets1,677 items0.8130.79 to 0.830.7950.78 to 0.810.7750.76 to 0.800.842 (highest on this suite)0.7810.7750.6940.701
Long inputs1,147 items0.8460.82 to 0.870.8150.79 to 0.840.853 (highest on this suite)0.83 to 0.870.8450.8280.4920.6340.758
Multilingual1,812 items0.801 (highest on this suite)0.78 to 0.820.7810.76 to 0.800.7820.76 to 0.800.7600.7370.8000.7560.727
Rules and policies1,759 items0.883 (highest on this suite)0.87 to 0.900.8660.85 to 0.880.8310.81 to 0.850.7980.7520.8450.7780.704
Prompt injection798 items0.7680.74 to 0.800.769 (highest on this suite)0.74 to 0.800.6000.56 to 0.630.5820.6480.3180.3270.652

Cloudflare Clef-flash 9B and Strands Decider 2B are in the mean above; their per-suite results are not published.

Text and images, one request

Two image suites: questions about real photos, and statements judged from rendered scenes. DecisionNode reads images and text together and answers through the same question types. TypeSafe Jev reads text only, so it is shown as not supported rather than as zero.

Photos

Choice and truth questions about real photos. 1,668 items.

DecisionNode
0.945
DecisionNode Flash
0.934
TypeSafe Jev 1.13
not supported, reads text only
Cloudflare Clef 27B
0.882
Cloudflare Clef-flash 9B
0.905
pplx-decider-27b
0.891
H2O-Lightning-4B
0.890

Not run on the image suites: Cygnet (Gemma 4 12B), Kev-4B and Strands Decider 2B.

Rendered scenes

How likely a statement is, judged from a computer-rendered scene. 450 items.

DecisionNode
0.762
DecisionNode Flash
0.680
TypeSafe Jev 1.13
not supported, reads text only
Cloudflare Clef 27B
0.783
Cloudflare Clef-flash 9B
0.742
pplx-decider-27b
0.809
H2O-Lightning-4B
0.684

Not run on the image suites: Cygnet (Gemma 4 12B), Kev-4B and Strands Decider 2B.

Places, video and drone views

Readings only our two models were measured on: where things are in a photo, questions about video and distances from drone views with Depth Vision. No other system on this board was run on them, so none is shown here. TypeSafe Jev reads text only.

One place in a photo

Point to a named thing, or draw a box around it. A point counts when it lands on the thing; a box counts when its overlap with the true box (IoU) is 0.5 or more. 300 photos.

Point on the thing
DecisionNode0.960DecisionNode Flash0.950
Box hit, IoU 0.5 or more
DecisionNode0.927DecisionNode Flash0.897
Box overlap, mean IoU
DecisionNode0.849DecisionNode Flash0.812

Every match in a photo

Find every match of a named thing, as points and as boxes. Precision is the share of returned places that are right; recall, the share of the real ones found. 197 photos.

Points, precision
DecisionNode0.721DecisionNode Flash0.688
Points, recall
DecisionNode0.672DecisionNode Flash0.663
Boxes, precision
DecisionNode0.692DecisionNode Flash0.646
Boxes, recall
DecisionNode0.547DecisionNode Flash0.522

Video

Choice, score and truth questions about short clips sent as frames, in four kinds of footage. 1,381 questions.

All questions
DecisionNode0.71595% interval 0.69 to 0.74DecisionNode Flash0.71895% interval 0.69 to 0.74
General scenes461 questions
DecisionNode0.86695% interval 0.83 to 0.89DecisionNode Flash0.84695% interval 0.81 to 0.88
Factory lines362 questions
DecisionNode0.71595% interval 0.67 to 0.76DecisionNode Flash0.74995% interval 0.71 to 0.79
Security cameras334 questions
DecisionNode0.64795% interval 0.60 to 0.70DecisionNode Flash0.66295% interval 0.61 to 0.71
Traffic224 questions
DecisionNode0.50995% interval 0.43 to 0.59DecisionNode Flash0.48795% interval 0.41 to 0.57

Drone views, with Depth Vision

Questions about pictures taken from drones: how far to the first obstacle ahead, whether the path is clear, which way is most open, how urgent it is to stop, and how far away the middle of the picture is. A distance counts when it is within 20% of the truth, or half a meter. Most questions are about rendered flights; 263 are about real drone photos. Each kind below is read with Depth Vision. 1,924 questions.

All questions, with Depth Vision
DecisionNode0.584DecisionNode Flash0.563
All questions, without it
DecisionNode0.525DecisionNode Flash0.524
Is the path clear for a given distance?400 questions
DecisionNode0.812DecisionNode Flash0.800
Which way is most open?176 questions
DecisionNode0.784DecisionNode Flash0.750
How urgent is it to stop?374 questions
DecisionNode0.802DecisionNode Flash0.775
How far to the obstacle ahead?338 questions
DecisionNode0.482DecisionNode Flash0.429
How far is the middle of the picture?636 questions
DecisionNode0.310DecisionNode Flash0.310

Real photos taken from high up are the hard part: there, most distance answers miss. Depth Vision's distances are estimates from one picture, for decisions and planning, never a safety stop.

Answered while a chat model is still typing

A chat LLM writes its answer token by token: GPT-5.6 Luna took 1,300 ms for a short decision, timed end to end from one client. DecisionNode reads the input once and answers in one pass: 19 ms end to end on DecisionNode Flash, 5 ms of it on our side.

up to 68×faster than a chat LLM on the same decision, end to end, on DecisionNode Flash

1 ms10 ms100 ms1,000 ms
  1. DecisionNode Flash19 ms end to end, 5 ms on our side
  2. GPT-5.6 Luna1,300 ms end to end

milliseconds end to end, log scale

End to end from one client, network included, one short decision, against GPT-5.6 Luna. October 2026. DecisionNode's latency is not published yet.

Accuracy is half of it. Calibration is the other half.

Every Choice and Score answer carries a confidence, and every Truth answer is the probability that a statement is true. Calibrated means those numbers mean what they say: of all the answers given at 0.9, about nine in ten are right.

That is what lets your code act alone. Set a threshold once and every answer above it is acted on instantly; below it, your code takes the safe path or re-checks with the full model. Calibration results per question type will be published with the full run.

How confidence works

  1. right
  2. right
  3. right
  4. right
  5. right
  6. right
  7. right
  8. right
  9. right
  10. wrong
Ten answers given at 0.90. Calibrated means about nine are right.an illustration of the definition, not a measurement

How we measured

15 text suites, 25,134 items, development splits. 95% intervals for our models and Jev from a bootstrap that resamples whole inputs. Jev was called through its official API; open models ran on our GPUs as their authors serve them. October 2026.

The suites

Held-out decisions636
Typed decisions on tasks kept out of development.
Public decision set117
A small public set of choice, score and truth questions.
Transfer tasks730
Decisions phrased the way other decision APIs phrase them.
Web tasks1,106
Pick the right element or action on a real web page.
Knowledge (MMLU-Pro)1,479
Hard multiple-choice questions that need world knowledge.
Short classifications178
One-line inputs sorted into a few labels.
Open decision tasks5,011
A broad public mix of typed decision tasks.
Typed decisions955
Mixed choice, score and truth questions in one request.
New task families3,629
Task families the model never saw while it was built.
Familiar task families4,100
New items from task families the model knows.
Large option sets1,677
Choices with dozens of options to pick from.
Long inputs1,147
Long threads and documents with the answer deep inside.
Multilingual1,812
The same decisions asked in many languages.
Rules and policies1,759
Apply a written policy to a case and decide.
Prompt injection798
Inputs that try to talk the model into the wrong answer.
Photos1,668, image
Choice and truth questions about real photos.
Rendered scenes450, image
How likely a statement is, judged from a computer-rendered scene.
  • Each suite is scored as accuracy: the share of items where the answer matches the reference. Intervals are 95%, from a bootstrap that resamples whole inputs, and are shown where they are published: for our models and TypeSafe Jev.
  • TypeSafe Jev 1.13 was called through its official API with its documented request shape. Open models ran on our GPUs, served the way their authors publish them.
  • Strands Decider 2B: its published training data contains items of two of our test sets, so its mean may read high.
  • Places, video and drone views were measured on our two models only, scored as each panel says. No other system here was run on them.

Information last reviewed: October 2026.

Every system here changes over time, ours included. Scores describe the versions measured on the date above; check each provider's current documentation before you choose.

DecisionNode is not affiliated with TypeSafe AI, Cloudflare, OpenAI or the other model publishers. Their names identify the systems measured.

Preliminary results, October 2026

Questions about the numbers

  • Why show the suites you lose?+

    Because a mean hides the shape. You should know that DecisionNode trails on knowledge-heavy questions and on task families it has never seen before you build on it, and that it leads on prompt injection, web tasks and written policies.

  • Are these numbers final?+

    No. They are preliminary results, october 2026, measured on the development splits of each suite, and the page says so wherever they appear.

  • Why are only your models on places, video and drone views?+

    Because no other system on this page was measured on them, so there is nothing to compare. TypeSafe Jev reads text only, according to its public documentation. The other systems were run on the suites above, not on these.

  • Why is Jev missing from the image suites?+

    TypeSafe Jev reads text only, according to its public documentation, so there is nothing to score. It is shown as not supported, not as zero.

  • Can I check the numbers on my own data?+

    Yes. Once you sign in, the playground runs any request against both models, and every answer is deterministic, so the same request gives the same answer every time you run it.

  • How fast is it next to the others?+

    DecisionNode Flash answers a short request in about 5 ms on our side and 19 ms end to end. GPT-5.6 Luna, timed from the same client on the same decision, took 1,300 ms, so DecisionNode Flash is up to 68× faster.

DecisionNodeDecisionNde

Add a decision engine with one prompt.

Paste into your coding agent.

start building

Your first decision in minutes.

Make a key, connect your project, branch on a typed answer. Prepaid, no sales call.

  • api all systems normalapi normal
  • output is always freeoutput always free
  • same request, same answersame request, same answer

DecisionNode is built and run by Bynn Intelligence, Inc.

We train the model and serve it on our own GPUs.

hello@bynn.com

product

  • Model
  • Inference
  • Benchmarks
  • Pricing
  • Open models

developers

  • Docs
  • Quickstart
  • Examples
  • Playground
  • Console
  • Status

company

  • Contact
  • Security
  • Report abuse
  • Dedicated
  • Affiliate program

legal

  • Terms
  • Acceptable use
  • Privacy
  • Data processing
  • Defence addendum
  • Cookies
  • Affiliate agreement

compliance

  • SOC 2, SOC 2 report, monitored continuously (opens the trust center in a new tab)
  • ISO 27001:2022, Information security management, certified (opens the trust center in a new tab)
  • ISO/IEC 42001:2023, AI management system, controls mapped to ISO/IEC 42001:2023; compliant, not certified (opens the trust center in a new tab)
  • GDPR, Processor terms under Article 28 in the Data Processing Addendum; a law, not a certification (opens the trust center in a new tab)
  • CCPA, Service-provider terms in the Data Processing Addendum; a law, not a certification (opens the trust center in a new tab)
  • PCI DSS SAQ A, Card data is handled by the payment provider and never stored by us; a self-assessment, not a certification (opens the trust center in a new tab)
  • NIST Cybersecurity Framework, Controls mapped to the NIST CSF; a framework, not a certification (opens the trust center in a new tab)
  • HIPAA (BAA available), Protected health information only under a signed Business Associate Agreement, available on request; HIPAA is not a certification (opens the trust center in a new tab)
  • EU AI Act, Controls mapped to the EU AI Act; a law, not a certification, and no determination that any customer use complies (opens the trust center in a new tab)

© 2026 Bynn Intelligence, Inc.

prices in USD per 1M input tokens