Decision model

Perplexity Decider 27B v1.3

1 question per request

Accuracy, all 2,198 questions
68.4%66.5% to 70.3%, 95% interval; 1,504 right
Answers kept at 99% confidence
276 of 2,1980 wrong; share right 98.6% to 100.0%, 95% interval
Cost, $ per 1,000 questionsper 1M questions
$0.081$81.18Basis and price under Cost
Latency
0.110 smedian request, 1 question per request

Back to the leaderboard.

Perplexity Decider 27B v1.2 · API and Perplexity Decider 27B v1.2 · API · with web, with Perplexity's web search off and on, are compared on Web grounding.

How we called it

Kind
Decision model
Access
Perplexity Decider server, self-hosted
Paid as
Our own hardware
Model ID
pplx-decider-v1.3
Variant
local
Parameters
27B dense, as the maker states it (counted from the released files: 26.09B).language model (no LM head) + vision encoder + 255-way readout; its locate-span adapter and head (0.47B), used only by its locate task, not counted; source: counted from the checkpoint's safetensors, perplexity-ai/pplx-decider-v1.3@9c2ebb2, huggingface.co/perplexity-ai/pplx-decider-v1.3; read 9 November 2026
Hardware
1 × RTX PRO 6000 Blackwell (96 GB); BF16
Note
Perplexity's released v1.3 checkpoint, the next version of Decider 27B v1.2 on Qwen3.8-27B, run on our hardware through the package it ships.
Confidence from
Option score
Run
1134, 14 November 2026

Settings

Reasoning
off
Request format
1 question per request
Requests at once
1
Temperature
not recorded
Run record: harness seed, harness commit and prompt version
Harness seed
0
Harness commit
015175ac1bde
Prompt version
254597b8be16

Results

QuestionsAccuracy95% intervalRight
All questions68.4%66.5% to 70.3%1,504 of 2,198
Reading89.0%86.4% to 91.2%584 of 656
Grammar59.7%57.2% to 62.1%920 of 1,542

Comparison group: zABCDE. Compare with another setup.

Accuracy by topic
TopicPartAccuracyRight
Word and phrase meaningsReading86.3%113 of 131
Sentence meaningsReading92.9%143 of 154
How a writer explains and develops ideasReading81.4%57 of 70
Topic and main point of a paragraphReading94.9%93 of 98
How a paragraph fits togetherReading82.1%87 of 106
Details that support a paragraph’s main pointReading93.8%91 of 97
Word types, such as nouns and verbsGrammar62.9%88 of 140
Groups of words built around a nounGrammar66.2%43 of 65
Verbs and their formsGrammar64.8%59 of 91
Verb forms used like nouns or modifiersGrammar69.4%50 of 72
Word endings and what they doGrammar59.5%47 of 79
How words are builtGrammar43.5%30 of 69
The roles of words and phrases in a sentenceGrammar60.7%88 of 145
How verbs show who acts and who receives the actionGrammar67.6%50 of 74
Types of sentencesGrammar62.4%53 of 85
Questions combining several grammar topicsGrammar50.5%111 of 220
Sound patterns and changesGrammar29.9%23 of 77
SpellingGrammar61.7%82 of 133
PunctuationGrammar61.5%72 of 117
Problems in sentence wordingGrammar70.9%124 of 175

English

The Text tab's 547 kept reading questions translated into English, asked with this setup's Text tab settings. Every setup: English reading; against Turkish: English and Turkish.

Right answers
490 of 547, first English run (1137)
Accuracy
89.6%
Non-hallucination with “I don't know”
8.8%: 5 “I don't know” of 57 not right (run 1139)
Answered with “I don't know”
99.1%: 542 of 547
Error rate with “I don't know”
9.6%: 52 of 542 answers
Cost, $ per 1,000 questionsper 1M questions, first English run
0.08382.63from GPU time
Latency, median request, first English run
0.115 s; 1 request at once
English run
1137
Machine, English runs
RTX PRO 6000
Date
14 November 2026
Request format
1 question per request
Reasoning
off
Prompt version
47f5776e71db
Translations, SHA-256
bf0d47cd385a93050c65906977e199ef7e45d5848dbbdef2bc8d124f4d5abf68

Confidence

Answers kept at 99% or more
276, of which 0 wrong (share right: 98.6% to 100.0%, 95% interval); 1,922 skipped, the model less than 99% sure
Calibration gap
5.0 points
AUROC, all questions
0.82 0.81 to 0.84
AUROC, reading
0.91 0.89 to 0.94
AUROC, grammar
0.76 0.74 to 0.78
Brier score
0.159
Accuracy of its surest answers
most confident fifth: 100.0% right; most confident half: 91.2% right; most confident four fifths: 76.8% right
Typical score of a right answer
90.4%
Typical score of a wrong answer
53.0%

Repeats

We did not repeat this setup.

Latency

Latency, median request
0.110 s (this run)
Latency, 90th percentile
0.145 s
Requests at once
1
Throughput
511.2 questions per minute of request time
Request time
4.3 minutes, over the requests whose answers were kept

Latency: the time from sending one request to receiving its answer, measured from our computer, network and any retries within the request included. Request time: the time at least one request was in flight, over the requests whose answers were kept. It is not the run's elapsed time: it leaves out requests that failed and any pause before a run was resumed.

Cost

Cost per 1,000 questionsper 1M questions
$0.081$81.18
Whole run
$0.18
Basis
Estimated from GPU time: the run's request time x the GPUs its server held x Runpod's on-demand price for an RTX PRO 6000, $2.49 per GPU-hour (read 5 November 2026). A cloud-equivalent estimate, not measured spending.
Price
$2.49 per GPU-hour, 1 GPU, for the run's request time (source, 2026-11-05)
Input tokens per question
372
Output tokens per question
0

Going first in front of an LLM

Replay of recorded answers, cutoff chosen on half the questions, averaged over the splits. Savings are estimates from API prices; the Method page explains how.

LLM behind itShare takenAccuracy changeSaving
Gemini 3.8 Flash high · chat23.6%−0.13 points20%
GPT-6 Astra high · chat22.4%−0.11 points21%
GPT-6.1 Sol high · chat26.0%−0.13 points21%
GPT-6 Astra low · chat22.4%−0.11 points21%
GPT-6 Astra low · chat · 1 per request22.3%−0.10 points22%
Gemini 3.8 Flash low · chat · 1 per request26.5%−0.13 points18%
GPT-6.1 Sol low · chat · 1 per request27.3%−0.17 points26%
GPT-6.1 Sol low · chat24.1%−0.12 points17%
Claude Opus 5.5 high · chat23.6%−0.13 points22%
GPT-6.1 Terra high · chat22.3%−0.10 points7%
Claude Opus 5.5 low · chat · 1 per request24.3%−0.16 points24%
Gemini 3.8 Flash low · chat25.5%−0.19 points-5%
GPT-6.1 Terra low · chat31.4%−0.16 points12%
Claude Opus 5.5 low · chat32.7%−0.26 points31%
Claude Sonnet 5.5 high · chat · 1 per request33.8%−0.26 points33%
Claude Sonnet 5.5 high · chat49.2%−0.05 points47%