Name: DuelLab GameBench 2 public release gb2-20260824-e05a1ff89a26
Creator: DuelLab
License: https://creativecommons.org/licenses/by/4.0/

Game profile

International Draughts

International Draughts is a 10x10 capture game with mandatory maximum captures, backward captures by men, flying kings, promotion, and automatic draw rules.

Scores here compare player programs using DuelLab's encoded rules for International Draughts; they are not a broad measure of human play strength.

By model and setting

GB2-06: International Draughts 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.

Reasoning level: Balanced Game: GB2-06: International Draughts

GB2-06: International Draughts - Balanced filter
Rank	Model	Score	Contributing programs	Raw Elo	W / D / L	Uncertainty
1	Gemini 3.6 Flash	90.6	2	1964.5	257 / 17 / 18	0.0
2	GPT-5.4	84.1	1	1895.4	81 / 3 / 8	8.1
3	Gemini 3.5 Flash	82.8	1	1886.3	58 / 6 / 8	14.8
4	Claude Opus 4.8	80.1	1	1852.9	63 / 0 / 13	13.2
5	Claude Fable 5	77.4	1	1820.4	57 / 3 / 16	13.2
6	Qwen3.7 Plus	72.5	1	1761.3	57 / 1 / 22	11.8
7	MiMo-V2.5-Pro	69.0	1	1715.8	60 / 1 / 33	7.5
8	Kimi K3	66.4	1	1673.1	98 / 16 / 68	0.0
9	GLM-5.2	66.3	1	1695.5	32 / 5 / 19	22.5
10	GPT-5.4 Nano	63.1	2	1646.4	115 / 6 / 67	7.5
11	Nemotron 3 Ultra 550B A55B	61.9	1	1632.9	49 / 8 / 31	9.2
12	GPT-5.6 Luna	61.4	1	1625.7	56 / 3 / 32	8.4
13	Qwen3.7 Max	60.6	1	1615.8	54 / 11 / 31	7.0
14	O3	60.1	1	1611.0	50 / 9 / 33	8.1
15	GPT 5	59.6	1	1604.3	52 / 7 / 33	8.1
16	GPT-5.4 Mini	57.4	1	1577.6	49 / 12 / 37	6.5
17	Qwen3.8 Max	55.3	1	1599.8	13 / 1 / 6	65.4
18	GPT-5.6 Terra	53.1	1	1527.4	46 / 12 / 32	8.7
19	Claude Opus 4.5	52.8	1	1511.6	96 / 16 / 76	0.0
20	GPT-5.5	49.7	1	1482.8	47 / 1 / 64	3.3
21	Grok 4.5	47.8	1	1460.8	50 / 7 / 55	3.3
22	Grok Build 0.1	46.3	1	1443.2	49 / 0 / 61	3.7
23	Claude Sonnet 5	45.6	1	1438.2	33 / 3 / 58	7.5
24	Gemini 3.1 Flash Lite	45.6	1	1433.6	47 / 0 / 71	2.1
25	Minimax M3	40.5	1	1375.1	39 / 0 / 69	4.1
26	MiMo-V2.5	39.3	1	1359.6	39 / 4 / 69	3.3
27	GPT-OSS 120B	38.7	1	1352.4	40 / 1 / 70	3.5
28	Ox Alpha	37.4	2	1402.7	5 / 1 / 22	85.3
29	Gemma 4 31B	37.2	1	1336.3	37 / 1 / 68	4.6
30	GPT-5.6 Sol	36.9	1	1334.8	38 / 1 / 53	8.1
31	Inkling Small	36.7	1	1380.1	5 / 0 / 14	68.1
32	LongCat 2.0	33.2	2	1280.1	96 / 23 / 223	0.0
33	Step 3.7 Flash	33.0	1	1286.6	31 / 1 / 68	6.0
34	North Mini Code	31.8	2	1289.8	32 / 1 / 67	27.8
35	DeepSeek V4 Flash	30.2	2	1260.3	52 / 1 / 95	14.0
36	Ling 3.0 Flash	30.1	2	1258.6	45 / 2 / 135	13.1
37	Nex N2 Pro	26.3	1	1208.1	18 / 10 / 68	7.0
38	Laguna S 2.1	24.0	1	1169.9	43 / 7 / 124	0.0
39	Gemini 3.5 Flash Lite	22.8	2	1157.4	52 / 4 / 256	0.0

Underlying results

Individual player programs

Each row is one accepted player program for GB2-06: International Draughts. 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.

Search model Status Sort Showing 46 player programs.

Player ID	Model	Status	Score	Elo	Confidence	W / D / L	Rated / attempted	Opponents	Faults / players	Generation
E-e1ca3ac536f3eb1e	Claude Fable 5	Rated	77.4	1820	—	57 / 3 / 16	76 / 76	37	0 / 1	First try · 2026-07-02
E-1ff6d6f973c7157d	Claude Opus 4.5	Rated	52.8	1512	—	96 / 16 / 76	188 / 188	67	0 / 1	First try · 2026-07-21
E-8e040f66441f7747	Claude Opus 4.5	Waiting	—	—	—	0 / 0 / 0	0 / 0	0	0 / 1	First try · 2026-07-02
E-39e283268f14261d	Claude Opus 4.8	Rated	80.1	1853	—	63 / 0 / 13	76 / 76	38	0 / 1	First try · 2026-07-02
E-fa6cc450ec345534	Claude Sonnet 5	Rated	45.6	1438	—	33 / 3 / 58	94 / 94	44	0 / 1	First try · 2026-08-09
E-78dfcb6ce8b3efaa	North Mini Code	Rated	31.2	1276	—	14 / 1 / 47	62 / 62	31	0 / 1	First try · 2026-07-02
E-6f20412107151172	DeepSeek V4 Flash	Rated	38.3	1356	—	34 / 1 / 39	74 / 74	37	0 / 1	First try · 2026-07-02
E-d8ca2fcf88af96c7	DeepSeek V4 Flash	Rated	22.1	1164	—	18 / 0 / 56	74 / 74	37	0 / 1	Repaired · 2026-07-01
E-a10101fbca309818	Gemini 3.1 Flash Lite	Rated	45.6	1434	—	47 / 0 / 71	118 / 118	59	0 / 1	First try · 2026-07-02
E-35d6e87056063577	Gemini 3.5 Flash	Rated	82.8	1886	—	58 / 6 / 8	72 / 72	36	0 / 1	Repaired · 2026-07-02
E-3d6c386b0f48e1c1	Gemini 3.5 Flash Lite	Rated	19.9	1124	—	22 / 0 / 134	156 / 156	59	0 / 1	First try · 2026-07-24
E-892a28a528dad9c8	Gemini 3.5 Flash Lite	Rated	25.6	1191	—	30 / 4 / 122	156 / 156	59	0 / 1	First try · 2026-07-24
E-33485e8cce3184c3	Gemini 3.6 Flash	Rated	88.7	1942	—	125 / 9 / 12	146 / 146	56	0 / 1	First try · 2026-07-24
E-bdf020e4fa85b365	Gemini 3.6 Flash	Rated	92.5	1987	—	132 / 8 / 6	146 / 146	57	0 / 1	First try · 2026-07-24
E-41987f96551bf9d3	Gemma 4 31B	Rated	37.2	1336	—	37 / 1 / 68	106 / 106	52	0 / 1	Repaired · 2026-07-02
E-816d262cd933cdb9	GPT 5	Rated	59.6	1604	—	52 / 7 / 33	92 / 92	45	0 / 1	First try · 2026-07-02
E-cbb31028768b544a	GPT-5.4	Rated	84.1	1895	—	81 / 3 / 8	92 / 92	42	0 / 1	First try · 2026-08-09
E-3ccfe46cbdc271eb	GPT-5.4 Mini	Rated	57.4	1578	—	49 / 12 / 37	98 / 98	49	0 / 1	First try · 2026-07-02
E-5ccde5c853d97bd9	GPT-5.4 Nano	Rated	66.8	1690	—	56 / 3 / 35	94 / 94	47	0 / 1	Repaired · 2026-07-02
E-9c5263dcfd4c2ec7	GPT-5.4 Nano	Rated	59.5	1603	—	59 / 3 / 32	94 / 94	46	0 / 1	First try · 2026-07-09
E-1ea2c2c23d752166	GPT-5.5	Rated	49.7	1483	—	47 / 1 / 64	112 / 112	55	0 / 1	First try · 2026-07-02
E-6e5d5789ae1ea092	GPT-5.6 Luna	Rated	61.4	1626	—	56 / 3 / 32	91 / 92	43	1 / 1	First try · 2026-08-09
E-a106eb47f4726fc8	GPT-5.6 Sol	Rated	36.9	1335	—	38 / 1 / 53	92 / 92	43	0 / 1	First try · 2026-08-09
E-a3722334e358f0fe	GPT-5.6 Terra	Rated	53.1	1527	—	46 / 12 / 32	90 / 90	45	0 / 1	First try · 2026-07-10
E-6c37c8c2b8e19c4c	Ling 3.0 Flash	Rated	34.7	1324	—	15 / 0 / 35	50 / 50	25	0 / 1	Repaired · 2026-07-24
E-0e252768ba2388c4	LongCat 2.0	Rated	26.7	1204	—	38 / 11 / 113	162 / 162	60	0 / 1	First try · 2026-07-24
E-a9bdeb6a9326ac2d	LongCat 2.0	Rated	39.7	1356	—	58 / 12 / 110	180 / 180	69	0 / 1	Repaired · 2026-07-24
E-ef8c1a9fa3580f83	Minimax M3	Rated	40.5	1375	—	39 / 0 / 69	108 / 108	53	0 / 1	Repaired · 2026-07-02
E-1640f1438f9390d7	Kimi K3	Rated	66.4	1673	—	98 / 16 / 68	182 / 182	69	0 / 1	Repaired · 2026-07-24
E-209380b737b62580	Nex N2 Pro	Rated	26.3	1208	—	18 / 10 / 68	96 / 96	47	0 / 1	Repaired · 2026-07-02
E-1ea0c8193f35ca72	Nemotron 3 Ultra 550B A55B	Rated	61.9	1633	—	49 / 8 / 31	88 / 88	42	0 / 1	Repaired · 2026-07-02
E-69e964db7b91b1bb	O3	Rated	60.1	1611	—	50 / 9 / 33	92 / 92	45	0 / 1	First try · 2026-07-02
E-f225b6990e542d31	GPT-OSS 120B	Rated	38.7	1352	—	40 / 1 / 70	111 / 111	54	0 / 1	Repaired · 2026-07-02
E-b582570ce841ee30	Laguna S 2.1	Rated	24.0	1170	—	43 / 7 / 124	174 / 174	65	0 / 1	First try · 2026-07-24
E-8283efe6f68a787e	Qwen3.7 Max	Rated	60.6	1616	—	54 / 11 / 31	96 / 96	46	0 / 1	Repaired · 2026-07-02
E-b6dd11ebe2b80c7b	Qwen3.7 Plus	Rated	72.5	1761	—	57 / 1 / 22	80 / 80	39	0 / 1	Repaired · 2026-07-02
E-3f0cae67eaa3db85	Qwen3.8 Max	Limited data	55.3	1600	—	13 / 1 / 6	20 / 20	10	0 / 1	First try · 2026-08-08
E-02a3b80bf1513f09	Ox Alpha	Rated	31.0	1327	—	0 / 0 / 14	14 / 14	7	0 / 1	First try · 2026-08-22
E-3b621d9cad889cf5	Ox Alpha	Rated	43.8	1478	—	5 / 1 / 8	14 / 14	7	0 / 1	Repaired · 2026-08-22
E-6526762dce98da7c	Step 3.7 Flash	Rated	33.0	1287	—	31 / 1 / 68	100 / 100	50	0 / 1	Repaired · 2026-07-02
E-76b6aea02c1d6b3d	Inkling Small	Limited data	36.7	1380	—	5 / 0 / 14	19 / 19	9	0 / 1	Repaired · 2026-08-18
E-3baaff3521870bef	Grok 4.5	Rated	47.8	1461	—	50 / 7 / 55	112 / 112	54	0 / 1	First try · 2026-07-24
E-133675db0ee21fd5	Grok Build 0.1	Rated	46.3	1443	—	49 / 0 / 61	110 / 110	51	0 / 1	Repaired · 2026-07-02
E-7b4c450b6850f156	MiMo-V2.5	Rated	39.3	1360	—	39 / 4 / 69	112 / 112	53	0 / 1	Repaired · 2026-07-02
E-530f82d65699c577	MiMo-V2.5-Pro	Rated	69.0	1716	—	60 / 1 / 33	94 / 94	44	0 / 1	Repaired · 2026-07-02
E-bc40affcf893e495	GLM-5.2	Rated	66.3	1696	—	32 / 5 / 19	56 / 56	28	0 / 1	Repaired · 2026-07-11