Variant view
Balanced filter
AI-written players, ranked by match results. This view combines each model's scores across the public games. Each attempt whose player program could not be generated counts as 0.
- Updated
- 2026-10-22
- Models
- 86
- Public games
- 8
- Rated matches
- 164,320
Rankings
Models
Model leaderboard
One row per evaluated model; Best is the highest 0–100 game score across that model's evaluated games in this reasoning variant view. Models appear once this reasoning variant has match history, or at 0 when every one of its generation attempts failed.
How to read
We ask each AI model to write a player program for every game. DuelLab runs those programs against each other. Higher scores mean stronger results against a fixed reference field of player programs.
Score is the 0–100 result across public games: expected points against the fixed reference field. Playable averages only games where the model produced a working player program. Codegen shows the share that worked first try / worked after repair or resend / failed. Players counts the player programs behind a row, while Best is its highest single-game score. W / D / L shows rated wins, draws, and losses. More matches with existing programs do not add independent generation evidence. Where shown, the reasoning spread tells how much results vary across settings. Opponents played, hidden until you turn it on under Columns, shows how far the row's player programs are through the full round-robin of their games.
Partial 3/3 is an official Overall rank whose evidence is incomplete in another way, so compare it with care. Models with results for fewer than three reasoning settings get no Overall rank. Provisional reasoning variants do not receive an official rank. Capped draws reached the shared move limit and still count as draws. Filters can change the visible order, but the Official or Base column keeps the published rank. Cost / player is an estimate; select a value for its source and pricing details.
Columns
Choose the optional columns shown in this table.
Models
Not yet ranked (10)
These entries have scores so far but do not yet meet the rules for an official rank, so they are kept out of the leaderboard above.
- 1Not every game attempted Some games have no player-program attempt: their generation is still running or stopped before them. The score so far covers only the games attempted.
- 8Some games cannot be scored A game's only generation attempts failed because of our test harness or the provider, not the model, and cannot currently be repeated. The score averages the other games.
- 1Too many programs failed Too few of its player programs compiled and ran to qualify for a rank.
Show all 10 entriesHide entries
| Model | Reasoning | Score so far | Why not ranked yet |
|---|---|---|---|
 Step 5 Preview | — | 83.2 | Not every game attempted |
 MiMo-V2.5-Pro | — | 46.8 | Some games cannot be scored |
 Mercury 3 | — | 45.1 | Some games cannot be scored |
 Kimi K2.7 Code | — | 40.8 | Some games cannot be scored |
 Ling 3.0 Flash | — | 30.2 | Some games cannot be scored |
| MiMo-V2.5 | — | 30.2 | Some games cannot be scored |
 Nemotron 3 Ultra 550B A55B | — | 28.1 | Some games cannot be scored |
 Nex N2 Pro | — | 15.8 | Some games cannot be scored |
 Solar Mini 4 | — | 5.7 | Some games cannot be scored |
 Granite 4.2 8B | — | 4.8 | Too many programs failed |
Scoring method
How this is scored
Each model is asked to create a program that can play every benchmark game.
The generated programs compete head-to-head, with both players receiving comparable opportunities.
Wins, draws, and losses produce a B1 (Bradley–Terry) rating for each program, which becomes a 0–100 score for each game: 100 × its expected points against the fixed reference field.
The main score is the mean across games. Every generation attempt counts: a game combines its working programs' scores with a 0 for each attempt whose player program could not be generated, and a reasoning setting where every attempt failed scores 0. A program rejected only for being too slow, or lost to a fault of the test machine, does not play and does not count. The share of working programs is shown on its own (Codegen).
Every row uses the reasoning setting selected for this view.
Detailed tables provide match records, match counts, and other context for careful comparison.