Skip to main content

PerplexityDecision modelActive

pplx-decider-v1-27b

PopularityRank 13 of 39 decision models

37out of 100new calculation

based on two data groups, mostly search, range 27 to 46

Summary

pplx-decider-v1-27b is Perplexity’s 27-billion-parameter checkpoint for decisions over a state. The Decisions API accepts text, JSON, and images, then asks through a fixed schema for Choice, Noul, or Score. The result contains probabilities and confidence instead of a free-form answer.

It is based on Qwen3.8-27B. You can call the model through the Perplexity Decisions API or load its Apache-2.0 weights locally. The API limits a request to under 262,144 tokens including state, images, and questions. It costs $0.04 per 1M input tokens, with no output-token price or additional per-request fee.

GraduallyBench Decision

pplx-decider-v1-27b decides 58 of 60 test cases correctly. The median of the 33 tested decision models is 60.0 %.

Accuracy

96.7 %

58 of 60 test cases

Robustness

96.7 %

29 of 30 reworded variants

Unnecessary reviews

0.0 %

0 of 52 cases without a required review

Response time

1.6 s

median, 95 % under 2.1 s, local, GPU

Accuracy compared

Share of correct decisions in percent, higher is better

20 of 33 models

pplx-decider-v1-27b96.7 %
Jev93.3 %
Kev 9B93.3 %
Clef91.7 %
Clef-Flash85.0 %
Decider 4B85.0 %
d1-2B85.0 %
Lev81.7 %
d1-Omni-2B81.7 %
Kev 4B78.3 %
Jeff 170.0 %
Decider 2B66.7 %
Kev 0.8B50.0 %

As of 11/10/2026. Source: GraduallyBench Decision

Accuracy by task familyThe tick marks the median of all 33 tested models.
Triage100.0 %12 / 12, median 58.3 %
Routing100.0 %12 / 12, median 83.3 %
Constraints100.0 %12 / 12, median 33.3 %
Tables83.3 %10 / 12, median 50.0 %
Evidence100.0 %12 / 12, median 75.0 %

pplx-decider-v1-27b

Benchmark run
Show test details

Decisions

Local · German, English

Execution Apple Metal. Precision BF16.

130 valid outputs from 130 attempted and 130 planned test slots.

Measured duration p50 / p95 1,616.5 / 2,101 ms (130 cases). Local GPU. Quantile calculation not documented.

Automatic decision accuracy
96.67 %(58 / 60 planned cases)German, English
Automatic decision accuracy, Triage
100 %(12 / 12 planned cases)German, English
Automatic decision accuracy, Routing
100 %(12 / 12 planned cases)German, English
Automatic decision accuracy, Selection with constraints
100 %(12 / 12 planned cases)German, English
Automatic decision accuracy, Numbers and tables
83.33 %(10 / 12 planned cases)German, English
Automatic decision accuracy, Sufficient evidence
100 %(12 / 12 planned cases)German, English
Correct robustness variants
96.67 %(29 / 30 planned cases)German, English
Correct development cases
100 %(40 / 40 planned cases)German, English
Unnecessary reviews
0 %(0 / 52 assessable cases without a required review)German, English
Missed required reviews
0 %(0 / 8 assessable cases requiring review)German, English

Response times are comparable only within the same environment and form no shared speed ranking. Where used, the test references were not independently reviewed by humans. Test coverage and methods differ by model. Each metric stands on its own; there is no combined overall score across tests. Data compiled on 11/10/2026. How we test · Download all Decision and Audio results as JSON

All benchmark results

Every published result with its source, test setup and comparison models.

Show all 16 published benchmark results

Page 1 of 2

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

WinoGrande: accuracy (Show measurement, test conditions, and source)
Source value
83.3
Metric
accuracy
Unit
%
Category
WinoGrande
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
FinancialPhraseBank: accuracy (Show measurement, test conditions, and source)
Source value
84.18
Metric
accuracy
Unit
%
Category
FinancialPhraseBank
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
RAGTruth: accuracy (Show measurement, test conditions, and source)
Source value
88.8
Metric
accuracy
Unit
%
Category
RAGTruth
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
JudgeBench: accuracy (Show measurement, test conditions, and source)
Source value
78.29
Metric
accuracy
Unit
%
Category
JudgeBench
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
BBH: accuracy (Show measurement, test conditions, and source)
Source value
82.8
Metric
accuracy
Unit
%
Category
BBH
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
JevBench public hard: accuracy (Show measurement, test conditions, and source)
Source value
70.3
Metric
accuracy
Unit
%
Category
JevBench public hard
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
TabFact: accuracy (Show measurement, test conditions, and source)
Source value
90.6
Metric
accuracy
Unit
%
Category
TabFact
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
ContractNLI: accuracy (Show measurement, test conditions, and source)
Source value
80.78
Metric
accuracy
Unit
%
Category
ContractNLI
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
Circa: accuracy (Show measurement, test conditions, and source)
Source value
89.2
Metric
accuracy
Unit
%
Category
Circa
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
Belebele: accuracy (Show measurement, test conditions, and source)
Source value
94
Metric
accuracy
Unit
%
Category
Belebele
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
TruthfulQA binary: accuracy (Show measurement, test conditions, and source)
Source value
85.4
Metric
accuracy
Unit
%
Category
TruthfulQA binary
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
active
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API
Source Perplexity Decider model card
Overall: Published aggregate (Show measurement, test conditions, and source)
Source value
85.71
Metric
Published aggregate
Unit
%
Category
Overall
Direction
Higher is better
Source type
vendor-reported
Evaluator
Perplexity Decider model card
Status
disputed
Retrieved at
2026-10-04
Methodology
Vendor-reported result. Run=Perplexity API; weighting=Not disclosed in model card
Context
Perplexity publishes 85.71%; its aggregation or weighting is undisclosed and differs from the arithmetic mean of the eleven printed rows (84.33%). The aggregate is not recomputed.
Source Perplexity Decider model card

Page 1 of 2

Popularity compared

How often the model is searched for, used and mentioned, compared with every model of its category.

Popularity index

0 to 100, higher is more popular

20 of 42 models

As of November 2026 · Source: gradually.ai popularity index

Over time

Weekly values from October 18, 2026 to November 20, 2026

pplx-decider-v1-27bAutoTrust JEV-27B 39Decider 4B 39pplx-decider-v1.1-27b 37Jeff 1 34

Method changes in the timeline, a jump at these points is no market event:

  1. 1November 1, 2026: Value only with search or usage data, 30-day attention
  2. 2November 1, 2026: Method version 4 (common scale, shrinkage)
  3. 3November 9, 2026: New topics measurement V20 (daily quorum, short forms)

What is gaining popularity right now, across models, vendors, harnesses and AI subscriptions, is collected in All AI trends.

All specifications

Architecture

Model class
Multimodal decision model, not a text generator
Source: Model class
Checkpoint
perplexity-ai/pplx-decider-v1-27b
Source: Checkpoint
Base model
Qwen3.8-27B
Source: Base model

Input and output

Input
Text, JSON, and images as state
Source: Input

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking
Unknown
no source
Watermarking technique
Not known
no source

Licence and access

Access
Perplexity Decisions API and local weights
Source: Access
License
Apache-2.0
Source: License
Shutdown
None announced
no source

API

API model ID
pplx-decider-v1-27b
Source: API model ID

Further details

Output schema
Noul, Choice, and Score with probabilities and confidence
Source: Output schema
API input limit
Under 262,144 tokens per request, including state, images, and questions
Source: API input limit

Prices

API price
$0.04 per 1M input tokens; output is free with no additional per-request feeSource

Sources and data date

Show all 10 sources

Benchmark sources at a glance

Which independent evaluators tested this model. This is coverage, not a quality ranking.

No active result from the reviewed external evaluators can currently be attributed to this model. Other source evidence remains available below.

Without vendor claims and historical or disputed results; those stay labelled in the single values.

TypeEvidence and data date
Research statusResearch date October 4, 2026. 4 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Unresolved

published-overall-aggregation

Note

The published Overall is not the arithmetic mean of the printed rows; weighting or source correction remains undisclosed.

Not found in the inspected sources

Exact model result in the public Decision Index 0.4.1 panel

Note

This index edition has no row that can be mapped to the profile repository/checkpoint exactly. This does not imply the model lacks results in other benchmarks.

Additional sourcePerplexity Decisions API (retrieved October 3, 2026) · Perplexity Decisions API quickstart · Editorial description reviewed October 3, 2026
Additional sourcepplx-decider model card (retrieved October 3, 2026; October 4, 2026) · pplx-decider model card · Editorial description reviewed October 3, 2026
Additional sourceStreamDecisionBench, Hosted APIs table (retrieved October 22, 2026)
Additional sourceRequesty decision-model benchmark (retrieved October 23, 2026)
Additional sourceOpenRouter: Popularity index (derived), CC BY 4.0
Additional sourceVercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional sourceOllama: cumulative downloads for the popularity index
Additional sourceReplicate: cumulative runs for the popularity index
Additional sourceGoogle Trending Now: limited feed selection for Germany and the US
Additional sourceCreative Commons Attribution 4.0