Mercury Decide
Inception · inception/mercury-decide · added 20 Oct 2026
Mercury Decide by Inception ranks #5 of 12 AI decision models overall on DecisionBench: #9 of 12 on Sudoku, #3 of 12 on Triage, #6 of 12 on Robustness. View on OpenRouter.
- Overall
- #5 of 12
- Average rank
- 6.0
- Sudoku
- #9 of 12
- Triage
- #3 of 12
- Robustness
- #6 of 12
Sudoku Logical reasoning
Leaderboard →- Rank
- #9 of 12
- Accuracy
- 75.7%
- Solved
- 20/20
- Mistakes / puzzle
- 16.1
- Solve time / puzzle
- 18s
- Cost / puzzle
- $0.003594
By difficulty
| Difficulty | Rank | Accuracy | Solved | Mistakes | Solve time | Cost |
|---|---|---|---|---|---|---|
| Easy | #11 | 84.4% | 5/5 | 7.6 | 13s | $0.001981 |
| Medium | #10 | 77.0% | 5/5 | 14.6 | 16s | $0.003395 |
| Hard | #7 | 76.3% | 5/5 | 16.8 | 18s | $0.003851 |
| Expert | #9 | 69.1% | 5/5 | 25.6 | 25s | $0.005149 |
Every race (20)
| Puzzle | Mistakes | Calls | Solve time | Cost | Status |
|---|---|---|---|---|---|
| easy-01 | 9 | 21 | 18s | $0.002350 | Solved |
| easy-02 | 10 | 22 | 11s | $0.002377 | Solved |
| easy-03 | 5 | 17 | 7.7s | $0.001584 | Solved |
| easy-04 | 6 | 17 | 8.0s | $0.001681 | Solved |
| easy-05 | 8 | 19 | 23s | $0.001915 | Solved |
| medium-01 | 22 | 34 | 16s | $0.004255 | Solved |
| medium-02 | 14 | 27 | 14s | $0.003269 | Solved |
| medium-03 | 14 | 27 | 23s | $0.003336 | Solved |
| medium-04 | 16 | 29 | 15s | $0.003552 | Solved |
| medium-05 | 7 | 22 | 11s | $0.002565 | Solved |
| hard-01 | 18 | 32 | 16s | $0.004126 | Solved |
| hard-02 | 13 | 27 | 19s | $0.003204 | Solved |
| hard-03 | 13 | 29 | 16s | $0.003618 | Solved |
| hard-04 | 25 | 36 | 24s | $0.004828 | Solved |
| hard-05 | 15 | 28 | 14s | $0.003477 | Solved |
| expert-01 | 25 | 36 | 21s | $0.004601 | Solved |
| expert-02 | 27 | 40 | 33s | $0.005376 | Solved |
| expert-03 | 23 | 39 | 22s | $0.005157 | Solved |
| expert-04 | 21 | 34 | 24s | $0.004390 | Solved |
| expert-05 | 32 | 45 | 26s | $0.006221 | Solved |
Triage Following written rules
Leaderboard →- Rank
- #3 of 12
- Accuracy
- 98.3%
- Team
- 100.0%
- Refund
- 100.0%
- Urgency
- 95.0%
- Confidence gap
- 0.014
60 tickets · 66.7% of mistakes made while at least 90% sure · $0.0159 per 1,000 tickets
Robustness Consistency under pressure
Leaderboard →- Rank
- #6 of 12
- Score
- 99.3%
- Same if re-asked
- 98.3%
- Changed by reorder
- 0.0%
- In a list of 40
- 98% / 100%
- Obeyed planted text
- 0.0%
20 tickets × 6 checks · changed by rename 0.0% · accuracy with planted text 98.3%