Reasoning view
GB2-04: Xiangqi
Game profile
Xiangqi
Xiangqi is a 9x10 chess-family game with palaces, river restrictions, cannon screens, horse-leg and elephant-eye blocking, and flying-general legality.
Scores here compare player programs using DuelLab's encoded rules for Xiangqi; they are not a broad measure of human play strength.
By model and setting
GB2-04: Xiangqi by model and setting
Player programs are ranked by relative per-game score (0–100). Raw rating is shown as an advanced metric, alongside wins, draws, losses, and a per-game uncertainty index.
| Rank | Model | Score | Contributing programs | Raw Elo | W / D / L | Uncertainty |
|---|---|---|---|---|---|---|
| 1 | 90.1 | 1 | 1805.9 | 64 / 11 / 3 | 12.5 | |
| 2 | 56.1 | 1 | 1589.2 | 27 / 27 / 14 | 16.4 | |
| 3 | 55.1 | 1 | 1578.0 | 24 / 54 / 6 | 10.5 | |
| 4 | 51.3 | 2 | 1553.2 | 49 / 80 / 45 | 9.6 | |
| 5 | 50.9 | 1 | 1553.5 | 16 / 43 / 17 | 13.2 | |
| 6 | 50.9 | 1 | 1550.6 | 26 / 47 / 13 | 9.9 | |
| 7 | 50.7 | 1 | 1552.4 | 19 / 49 / 8 | 13.2 | |
| 8 | 46.2 | 1 | 1519.8 | 10 / 69 / 9 | 9.2 | |
| 9 | 44.8 | 1 | 1511.1 | 13 / 66 / 9 | 9.2 | |
| 10 | 43.8 | 1 | 1505.6 | 16 / 54 / 14 | 10.5 | |
| 11 | 43.3 | 1 | 1504.1 | 7 / 64 / 5 | 13.2 | |
| 12 | 42.2 | 1 | 1487.1 | 22 / 88 / 17 | 0.4 | |
| 13 | 42.0 | 1 | 1491.4 | 16 / 58 / 20 | 7.5 | |
| 14 | 41.7 | 1 | 1494.8 | 7 / 63 / 4 | 14.0 | |
| 15 | 41.5 | 1 | 1497.4 | 7 / 46 / 9 | 19.2 | |
| 16 | 41.3 | 1 | 1485.6 | 12 / 75 / 13 | 6.0 | |
| 17 | 40.8 | 1 | 1485.9 | 5 / 73 / 6 | 10.5 | |
| 18 | 39.5 | 1 | 1475.5 | 10 / 68 / 15 | 7.8 | |
| 19 | 39.5 | 1 | 1475.1 | 20 / 58 / 16 | 7.5 | |
| 20 | 39.3 | 1 | 1474.5 | 5 / 77 / 10 | 8.1 | |
| 21 | 39.2 | 2 | 1470.0 | 18 / 183 / 27 | 3.2 | |
| 22 | 38.8 | 1 | 1475.0 | 14 / 46 / 16 | 13.2 | |
| 23 | 37.8 | 3 | 1466.6 | 16 / 185 / 49 | 10.4 | |
| 24 | 37.8 | 1 | 1468.1 | 8 / 65 / 5 | 12.5 | |
| 25 | 37.4 | 1 | 1457.8 | 7 / 90 / 19 | 2.5 | |
| 26 | 37.3 | 1 | 1470.7 | 4 / 49 / 9 | 19.2 | |
| 27 | 36.9 | 2 | 1453.8 | 30 / 172 / 37 | 2.1 | |
| 28 | 36.8 | 1 | 1460.3 | 2 / 75 / 7 | 10.5 | |
| 29 | 36.6 | 2 | 1474.1 | 7 / 75 / 12 | 29.4 | |
| 30 | 36.4 | 1 | 1458.7 | 9 / 60 / 11 | 11.8 | |
| 31 | 35.2 | 3 | 1503.2 | 9 / 30 / 9 | 77.5 | |
| 32 | 34.8 | 1 | 1449.1 | 9 / 58 / 11 | 12.5 | |
| 33 | 34.6 | 1 | 1445.2 | 8 / 68 / 12 | 9.2 | |
| 34 | 32.6 | 2 | 1437.5 | 13 / 125 / 26 | 16.2 | |
| 35 | 31.6 | 1 | 1470.8 | 2 / 13 / 5 | 65.4 | |
| 36 | 31.1 | 1 | 1477.0 | 3 / 9 / 4 | 77.5 | |
| 37 | 29.4 | 1 | 1414.4 | 9 / 40 / 27 | 13.2 | |
| 38 | 27.2 | 1 | 1398.8 | 8 / 41 / 33 | 11.1 | |
| 39 | 26.5 | 1 | 1383.3 | 12 / 90 / 46 | 0.0 | |
| 40 | 19.5 | 2 | 1340.8 | 19 / 147 / 88 | 1.2 |
Underlying results
Individual player programs
Each row is one accepted player program for GB2-04: Xiangqi. A model can appear more than once when it produced more than one program. Score is the normalized 0–100 result; Elo and W / D / L show the underlying play record. Confidence and evidence can be blank when too few compatible matches are available. Draws at the move limit count; one-sided model faults are listed separately.
| Player ID | Model | Status | Score | Elo | Confidence | W / D / L | Rated / attempted | Opponents | Capped draws | Faults / players | Generation |
|---|---|---|---|---|---|---|---|---|---|---|---|
| E-eb1f17ce19beffd0 | Claude Fable 5 | Rated | 56.1 | 1589 | — | 27 / 27 / 14 | 68 / 68 | 34 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-8c9e611d78967342 | Claude Opus 4.5 | Waiting | — | — | — | 0 / 0 / 0 | 0 / 0 | 0 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-cbae2963ded72d76 | Claude Opus 4.5 | Rated | 39.5 | 1476 | — | 10 / 68 / 15 | 93 / 94 | 46 | 0 | 0 / 1 | First try · 2026-07-21 |
| E-caf91ab5b69b1e64 | Claude Opus 4.8 | Rated | 39.3 | 1475 | — | 5 / 77 / 10 | 92 / 92 | 43 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-048d3dcf802043b4 | Claude Opus 5 | Rated | 51.1 | 1552 | — | 15 / 50 / 21 | 86 / 86 | 43 | 0 | 0 / 1 | First try · 2026-07-25 |
| E-d0f0a2b1c4bbddae | Claude Opus 5 | Rated | 51.6 | 1554 | — | 34 / 30 / 24 | 88 / 88 | 44 | 0 | 0 / 1 | First try · 2026-07-25 |
| E-7031fd4afe7acb91 | Claude Sonnet 5 | Rated | 38.8 | 1475 | — | 14 / 46 / 16 | 76 / 76 | 36 | 0 | 0 / 1 | First try · 2026-08-09 |
| E-1460cf3cd86be35c | North Mini Code | Rated | 37.3 | 1473 | — | 3 / 45 / 8 | 56 / 56 | 28 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-5d143d6301429619 | DeepSeek V4 Flash | Rated | 37.3 | 1471 | — | 4 / 49 / 9 | 62 / 62 | 31 | 0 | 0 / 1 | First try · 2026-07-01 |
| E-860c31323fa5eda7 | Gemini 3.1 Flash Lite | Rated | 41.7 | 1495 | — | 7 / 63 / 4 | 74 / 76 | 35 | 0 | 2 / 1 | Repaired · 2026-07-02 |
| E-32cfa6c9dfb1d16f | Gemini 3.5 Flash | Rated | 50.9 | 1551 | — | 26 / 47 / 13 | 86 / 86 | 41 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-5c270e2afa7b70f5 | Gemini 3.5 Flash Lite | Rated | 42.5 | 1494 | — | 10 / 78 / 11 | 99 / 102 | 50 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-6098de52ddfabfb1 | Gemini 3.5 Flash Lite | Rated | 35.9 | 1446 | — | 8 / 105 / 16 | 129 / 135 | 53 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-cefb1dec9ce36375 | Gemini 3.6 Flash | Rated | 30.8 | 1413 | — | 10 / 96 / 25 | 131 / 131 | 52 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-d109ab12db58a65f | Gemini 3.6 Flash | Rated | 43.0 | 1495 | — | 20 / 76 / 12 | 108 / 108 | 54 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-62ca2464f2212fc9 | Gemma 4 31B | Rated | 37.8 | 1468 | — | 8 / 65 / 5 | 78 / 80 | 34 | 0 | 2 / 1 | Repaired · 2026-07-02 |
| E-6f1aae863396df76 | GPT 5 | Rated | 43.8 | 1506 | — | 16 / 54 / 14 | 84 / 84 | 40 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-cae96a5e03c2613e | GPT-5.4 | Rated | 27.2 | 1399 | — | 8 / 41 / 33 | 82 / 82 | 37 | 0 | 0 / 1 | First try · 2026-08-09 |
| E-74f5bcc8ad3796a5 | GPT-5.4 Mini | Rated | 50.7 | 1552 | — | 19 / 49 / 8 | 76 / 76 | 38 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-a32cb67d48c62b3b | GPT-5.4 Nano | Rated | 55.1 | 1578 | — | 24 / 54 / 6 | 84 / 84 | 39 | 0 | 0 / 1 | Repaired · 2026-07-09 |
| E-9e7bdad81107b7e4 | GPT-5.5 | Rated | 39.5 | 1475 | — | 20 / 58 / 16 | 94 / 94 | 47 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-3c11110e415436c3 | GPT-5.6 Luna | Rated | 34.8 | 1449 | — | 9 / 58 / 11 | 78 / 78 | 36 | 0 | 0 / 1 | First try · 2026-08-09 |
| E-7f8aa62c2a56b424 | GPT-5.6 Sol | Rated | 90.1 | 1806 | — | 64 / 11 / 3 | 78 / 78 | 36 | 0 | 0 / 1 | First try · 2026-08-09 |
| E-89c18d144e85d89f | GPT-5.6 Terra | Rated | 44.8 | 1511 | — | 13 / 66 / 9 | 88 / 88 | 43 | 0 | 0 / 1 | First try · 2026-07-10 |
| E-20e94e7276184071 | Ling 3.0 Flash | Rated | 29.3 | 1428 | — | 3 / 32 / 9 | 44 / 44 | 22 | 0 | 0 / 1 | Repaired · 2026-07-24 |
| E-4ccf296e55b59be3 | LongCat 2.0 | Rated | 17.0 | 1326 | — | 14 / 57 / 45 | 116 / 139 | 49 | 0 | 23 / 1 | Repaired · 2026-07-24 |
| E-9c9e87bb30b3cb99 | LongCat 2.0 | Rated | 22.1 | 1356 | — | 5 / 90 / 43 | 138 / 138 | 53 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-57e6f935250459a2 | Muse Spark 1.2 | Rated | 29.4 | 1414 | — | 9 / 40 / 27 | 76 / 76 | 36 | 0 | 0 / 1 | First try · 2026-08-13 |
| E-c60620c3c72a6064 | Minimax M3 | Rated | 43.3 | 1504 | — | 7 / 64 / 5 | 76 / 76 | 37 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-eb97811a38950c20 | Kimi K2.7 Code | Rated | 34.6 | 1445 | — | 8 / 68 / 12 | 88 / 88 | 43 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-7c724729d138057b | Kimi K3 | Rated | 42.2 | 1487 | — | 22 / 88 / 17 | 127 / 131 | 50 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-7e92e17a121661f4 | O3 | Rated | 36.4 | 1459 | — | 9 / 60 / 11 | 80 / 80 | 38 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-f1288f2abfdbbfe4 | GPT-OSS 120B | Rated | 36.8 | 1460 | — | 2 / 75 / 7 | 84 / 84 | 40 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-e996fd38983e3b7f | Laguna S 2.1 | Rated | 26.5 | 1383 | — | 12 / 90 / 46 | 148 / 148 | 57 | 0 | 0 / 1 | Repaired · 2026-07-24 |
| E-ea08328e287de1ad | Qwen3.7 Max | Rated | 42.0 | 1491 | — | 16 / 58 / 20 | 94 / 94 | 42 | 0 | 0 / 1 | First try · 2026-07-02 |
| E-1f87c69eddf28f4e | Qwen3.8 Max | Limited data | 31.1 | 1477 | — | 3 / 9 / 4 | 16 / 16 | 8 | 0 | 0 / 1 | First try · 2026-08-08 |
| E-43ae1b9ca4bd7996 | Ox Alpha | Rated | 41.6 | 1545 | — | 6 / 8 / 2 | 16 / 16 | 8 | 0 | 0 / 1 | Repaired · 2026-08-22 |
| E-6139f660620fd73b | Ox Alpha | Rated | 34.5 | 1499 | — | 2 / 11 / 3 | 16 / 16 | 8 | 0 | 0 / 1 | First try · 2026-08-22 |
| E-be47453e36658247 | Ox Alpha | Rated | 29.3 | 1465 | — | 1 / 11 / 4 | 16 / 16 | 8 | 0 | 0 / 1 | First try · 2026-08-22 |
| E-25e56c6bfae33c62 | Step 3.7 Flash | Rated | 46.2 | 1520 | — | 10 / 69 / 9 | 88 / 88 | 42 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-fce01db31ecf26d2 | Inkling | Rated | 37.4 | 1458 | — | 7 / 90 / 19 | 116 / 116 | 44 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-20d371c6aae692a2 | Inkling Small | Limited data | 31.6 | 1471 | — | 2 / 13 / 5 | 20 / 20 | 10 | 0 | 0 / 1 | Repaired · 2026-08-09 |
| E-ba6a8e6d7fc9a48d | Grok 4.5 | Rated | 41.3 | 1486 | — | 12 / 75 / 13 | 100 / 100 | 49 | 0 | 0 / 1 | First try · 2026-07-24 |
| E-85934b32572399da | Grok Build 0.1 | Rated | 40.8 | 1486 | — | 5 / 73 / 6 | 84 / 84 | 40 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-00e6914e513b1656 | MiMo-V2.5 | Rated | 37.0 | 1463 | — | 7 / 55 / 16 | 78 / 78 | 37 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-23e877112cb95491 | MiMo-V2.5 | Rated | 35.2 | 1450 | — | 5 / 55 / 24 | 84 / 84 | 39 | 0 | 0 / 1 | Repaired · 2026-07-01 |
| E-2f7e15b8aa123825 | MiMo-V2.5 | Rated | 39.5 | 1476 | — | 4 / 75 / 9 | 88 / 88 | 41 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-ffdebde8a72ed049 | MiMo-V2.5-Pro | Rated | 50.9 | 1554 | — | 16 / 43 / 17 | 76 / 76 | 37 | 0 | 0 / 1 | Repaired · 2026-07-02 |
| E-912f31ad14c4bf0a | GLM-5.2 | Rated | 41.5 | 1497 | — | 7 / 46 / 9 | 62 / 62 | 31 | 0 | 0 / 1 | Repaired · 2026-07-11 |