Decider v1.2
Perplexity · perplexity/pplx-decider-v1.2 · added 20 Oct 2026
Decider v1.2 by Perplexity ranks #2 of 12 AI decision models overall on DecisionBench: #8 of 12 on Sudoku, #4 of 12 on Triage, #1 of 12 on Robustness. View on OpenRouter.
- Overall
- #2 of 12
- Average rank
- 4.3
- Sudoku
- #8 of 12
- Triage
- #4 of 12
- Robustness
- #1 of 12
Sudoku Logical reasoning
Leaderboard →- Rank
- #8 of 12
- Accuracy
- 77.3%
- Solved
- 20/20
- Mistakes / puzzle
- 14.8
- Solve time / puzzle
- 40s
- Cost / puzzle
- $0.002168
By difficulty
| Difficulty | Rank | Accuracy | Solved | Mistakes | Solve time | Cost |
|---|---|---|---|---|---|---|
| Easy | #8 | 88.4% | 5/5 | 5.4 | 14s | $0.001095 |
| Medium | #9 | 79.5% | 5/5 | 12.6 | 43s | $0.001899 |
| Hard | #10 | 71.2% | 5/5 | 21.8 | 48s | $0.002869 |
| Expert | #6 | 74.9% | 5/5 | 19.2 | 55s | $0.002811 |
Every race (20)
| Puzzle | Mistakes | Calls | Solve time | Cost | Status |
|---|---|---|---|---|---|
| easy-01 | 4 | 17 | 14s | $0.001083 | Solved |
| easy-02 | 3 | 15 | 10s | $0.000887 | Solved |
| easy-03 | 9 | 19 | 16s | $0.001180 | Solved |
| easy-04 | 10 | 22 | 18s | $0.001488 | Solved |
| easy-05 | 1 | 15 | 10s | $0.000839 | Solved |
| medium-01 | 16 | 28 | 27s | $0.002042 | Solved |
| medium-02 | 12 | 25 | 21s | $0.001727 | Solved |
| medium-03 | 14 | 27 | 75s | $0.002035 | Solved |
| medium-04 | 10 | 25 | 64s | $0.001829 | Solved |
| medium-05 | 11 | 25 | 27s | $0.001861 | Solved |
| hard-01 | 23 | 37 | 48s | $0.002989 | Solved |
| hard-02 | 19 | 31 | 32s | $0.002400 | Solved |
| hard-03 | 19 | 35 | 37s | $0.002769 | Solved |
| hard-04 | 22 | 36 | 41s | $0.002955 | Solved |
| hard-05 | 26 | 39 | 84s | $0.003231 | Solved |
| expert-01 | 24 | 40 | 85s | $0.003316 | Solved |
| expert-02 | 14 | 30 | 31s | $0.002341 | Solved |
| expert-03 | 16 | 31 | 32s | $0.002475 | Solved |
| expert-04 | 19 | 34 | 41s | $0.002790 | Solved |
| expert-05 | 23 | 37 | 88s | $0.003133 | Solved |
Triage Following written rules
Leaderboard →- Rank
- #4 of 12
- Accuracy
- 98.3%
- Team
- 100.0%
- Refund
- 100.0%
- Urgency
- 95.0%
- Confidence gap
- 0.015
60 tickets · 100.0% of mistakes made while at least 90% sure · $0.0141 per 1,000 tickets
Robustness Consistency under pressure
Leaderboard →- Rank
- #1 of 12
- Score
- 100.0%
- Same if re-asked
- 100.0%
- Changed by reorder
- 0.0%
- In a list of 40
- 100% / 100%
- Obeyed planted text
- 0.0%
20 tickets × 6 checks · changed by rename 0.0% · accuracy with planted text 100.0%