preference docket
Leaderboard
Community preference, one track at a time. Not correctness. Not intelligence. Picks and ties move Elo; skips don’t.
Challenge
v0| Rank | Model | Elo | Comparisons | Status |
|---|---|---|---|---|
| 1 | Kimi K2.7 Coding/kimi-k2-7-coding-default | 1512.00 | 1 | provisional |
| 2 | MiMo V2.5/mimo-v2-5-default | 1512.00 | 1 | provisional |
| 3 | Claude Sonnet 5 (max)/sonnet-5-max | 1512.00 | 1 | provisional |
| 4 | Grok 4.6 high/grok-4-6-high | 1488.00 | 1 | provisional |
| 5 | Nemotron 3.5 Lightning/nemotron-3-5-lightning-default | 1488.00 | 1 | provisional |
| 6 | Claude Sonnet 5 (high)/sonnet-5-high | 1488.00 | 1 | provisional |
Provisional: under 20 comparisons. A comparison is a pick or a tie. The prompt is on if posted, is on challenge. Scoring on methodology.