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

Game profile

Go

Go is a large-grid territory-control game with stone placement, captures, ko restrictions, passing, and final area scoring.

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

By model and setting

GB2-01: Go 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: Baseline Game: GB2-01: Go

GB2-01: Go - Baseline filter
Rank	Model	Score	Contributing programs	Raw Elo	W / D / L	Uncertainty
1	GPT-5.6 Sol	78.7	1	1827.9	68 / 0 / 20	9.2
2	GPT-5.4	73.2	1	1772.8	70 / 4 / 20	7.5
3	Hy3 Preview	72.9	1	1773.9	55 / 14 / 9	12.5
4	Claude Opus 5	66.1	2	1702.5	95 / 92 / 16	5.7
5	GPT 4.1	63.3	1	1678.2	55 / 5 / 31	8.4
6	Qwen3.7 Max	60.4	1	1652.4	41 / 23 / 16	11.8
7	GPT-5.4 Nano	56.1	2	1609.1	76 / 51 / 49	9.3
8	Claude Opus 4.8	55.5	1	1606.8	38 / 20 / 18	13.2
9	GPT-5.6 Terra	53.9	1	1589.3	26 / 30 / 24	11.8
10	Minimax M3	52.0	1	1564.9	59 / 21 / 32	3.3
11	Nemotron 3 Ultra 550B A55B	50.9	1	1564.6	25 / 13 / 30	16.4
12	GPT-5.5	48.5	1	1531.2	11 / 84 / 15	3.7
13	Laguna S 2.1	46.5	2	1512.7	78 / 50 / 82	4.9
14	MiMo-V2.5	46.5	3	1514.2	104 / 60 / 114	7.2
15	Kimi K3	46.3	1	1500.1	32 / 130 / 29	0.0
16	Grok 4.5	45.8	1	1494.9	37 / 126 / 34	0.0
17	Gemma 4 31B	44.0	1	1487.8	53 / 23 / 34	3.7
18	DeepSeek V4 Pro	43.6	1	1483.0	34 / 44 / 38	2.5
19	Claude Sonnet 5	43.2	1	1484.2	28 / 16 / 43	9.6
20	Mistral Medium 3.5	40.9	1	1459.8	40 / 9 / 47	7.0
21	Claude Opus 4.5	40.3	1	1452.1	15 / 71 / 22	4.1
22	GPT 4.1 Nano	39.7	1	1439.5	60 / 35 / 67	0.0
23	Inkling Small	39.7	1	1497.2	9 / 3 / 7	68.1
24	Ling 3.0 Flash	37.0	2	1414.6	91 / 95 / 108	0.0
25	MiMo-V2.5-Pro	36.9	1	1418.7	32 / 29 / 47	4.1
26	Gemini 3.5 Flash Lite	36.6	2	1408.7	111 / 74 / 157	0.0
27	LongCat 2.0	33.6	2	1386.4	64 / 78 / 88	2.7
28	GPT-5.4 Mini	32.4	1	1377.9	20 / 38 / 40	6.5
29	Ox Alpha	32.2	3	1438.8	12 / 3 / 27	85.3
30	Muse Spark 1.2	32.1	1	1376.4	9 / 37 / 44	8.7
31	GLM-5.2	28.7	2	1347.6	49 / 28 / 71	14.0
32	Qwen3.7 Plus	27.5	1	1330.7	18 / 25 / 51	7.5
33	Qwen3.8 Max	25.5	1	1367.2	2 / 1 / 13	77.5
34	North Mini Code	23.2	2	1287.6	21 / 96 / 89	5.7
35	Gemini 3.6 Flash	22.9	2	1276.0	77 / 74 / 183	0.0
36	GPT-5.6 Luna	17.9	1	1239.5	9 / 12 / 68	9.0
37	GPT-OSS 120B	17.7	1	1226.3	27 / 31 / 108	0.0
38	Inkling	12.6	2	1183.2	21 / 79 / 122	3.5
39	Gemini 3.1 Flash Lite	10.2	1	1168.7	6 / 24 / 44	14.0
40	O3	2.8	1	1082.9	5 / 42 / 111	0.0

Underlying results

Individual player programs

Each row is one accepted player program for GB2-01: Go. 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 56 player programs.

Player ID	Model	Status	Score	Elo	Confidence	W / D / L	Rated / attempted	Opponents	Capped draws	Faults / players	Generation
E-2e3eca63f7980fac	Claude Opus 4.5	Rated	40.3	1452	—	15 / 71 / 22	108 / 108	53	71	0 / 1	First try · 2026-07-02
E-757a6149e87d8b53	Claude Opus 4.8	Rated	55.5	1607	—	38 / 20 / 18	76 / 76	38	20	0 / 1	First try · 2026-07-02
E-6c30dc38efb07e9e	Claude Opus 5	Rated	49.8	1544	—	15 / 81 / 9	105 / 106	51	81	0 / 1	First try · 2026-07-25
E-865d68ab219fd7ee	Claude Opus 5	Rated	82.4	1861	—	80 / 11 / 7	98 / 98	47	11	0 / 1	First try · 2026-07-25
E-93cff8e6feb76995	Claude Sonnet 5	Rated	43.2	1484	—	28 / 16 / 43	87 / 87	41	16	0 / 1	First try · 2026-08-09
E-194759b182ce3244	North Mini Code	Rated	46.4	1509	—	19 / 75 / 24	118 / 118	54	75	0 / 1	First try · 2026-07-02
E-c309b463b29efb71	North Mini Code	Rated	0.0	1066	—	2 / 21 / 65	88 / 88	42	21	0 / 1	Repaired · 2026-07-11
E-e121525dd1846069	DeepSeek V4 Pro	Rated	43.6	1483	—	34 / 44 / 38	116 / 116	56	44	0 / 1	First try · 2026-07-02
E-cc0af0bfa6a9a29f	Gemini 3.1 Flash Lite	Rated	10.2	1169	—	6 / 24 / 44	74 / 74	37	24	0 / 1	Repaired · 2026-07-02
E-b33eea8b864e22bf	Gemini 3.5 Flash	Waiting	—	—	—	0 / 0 / 0	0 / 0	0	0	0 / 1	First try · 2026-07-21
E-21c0edac8ac36f2e	Gemini 3.5 Flash Lite	Rated	34.7	1390	—	62 / 39 / 71	172 / 172	66	39	0 / 1	First try · 2026-07-24
E-cf51875df5447bc2	Gemini 3.5 Flash Lite	Rated	38.5	1427	—	49 / 35 / 86	170 / 170	64	35	0 / 1	First try · 2026-07-24
E-e0c4f14c94c0a42a	Gemini 3.6 Flash	Rated	16.2	1211	—	33 / 43 / 92	168 / 168	63	43	0 / 1	First try · 2026-07-24
E-e9733de94d982ade	Gemini 3.6 Flash	Rated	29.6	1341	—	44 / 31 / 91	166 / 166	63	31	0 / 1	First try · 2026-07-24
E-caaa54d3c88e6570	Gemma 4 31B	Rated	44.0	1488	—	53 / 23 / 34	110 / 110	53	23	0 / 1	First try · 2026-07-02
E-f50ec6b60d5b85da	GPT 4.1	Rated	63.3	1678	—	55 / 5 / 31	91 / 92	45	5	1 / 1	First try · 2026-07-02
E-e117cfcfd3d0bf38	GPT 4.1 Nano	Rated	39.7	1440	—	60 / 35 / 67	162 / 162	61	35	0 / 1	First try · 2026-07-12
E-4370194a9beec334	GPT 5	Waiting	—	—	—	0 / 0 / 0	0 / 0	0	0	0 / 1	First try · 2026-07-21
E-ed2ff19b1858f2c4	GPT-5.4	Rated	73.2	1773	—	70 / 4 / 20	94 / 94	41	4	0 / 1	First try · 2026-08-09
E-ea1cb793f92b3cff	GPT-5.4 Mini	Rated	32.4	1378	—	20 / 38 / 40	98 / 98	48	38	0 / 1	First try · 2026-07-02
E-9eb27042bf28d67c	GPT-5.4 Nano	Rated	60.3	1651	—	44 / 19 / 19	82 / 82	40	19	0 / 1	First try · 2026-07-09
E-a8887794a1195b75	GPT-5.4 Nano	Rated	51.9	1567	—	32 / 32 / 30	94 / 100	46	32	0 / 1	Repaired · 2026-07-02
E-58178b9b32a94ca0	GPT-5.5	Rated	48.5	1531	—	11 / 84 / 15	110 / 110	54	84	0 / 1	First try · 2026-07-02
E-9502190155453416	GPT-5.6 Luna	Rated	17.9	1240	—	9 / 12 / 68	89 / 89	42	12	0 / 1	First try · 2026-08-09
E-72b9641e74736185	GPT-5.6 Sol	Rated	78.7	1828	—	68 / 0 / 20	88 / 88	41	0	0 / 1	First try · 2026-08-09
E-40742129939a44b5	GPT-5.6 Terra	Rated	53.9	1589	—	26 / 30 / 24	80 / 80	39	30	0 / 1	First try · 2026-07-10
E-3c618b88b6bdb039	Ling 3.0 Flash	Rated	30.2	1348	—	49 / 54 / 47	150 / 150	58	54	0 / 1	First try · 2026-07-24
E-f9af396559c3a1cd	Ling 3.0 Flash	Rated	43.8	1481	—	42 / 41 / 61	144 / 144	51	41	0 / 1	First try · 2026-07-24
E-9f5585b5fd96f2c3	LongCat 2.0	Rated	37.3	1423	—	34 / 32 / 44	110 / 110	52	32	0 / 1	First try · 2026-07-24
E-ed0f8e084958eebc	LongCat 2.0	Rated	29.9	1350	—	30 / 46 / 44	120 / 120	57	46	0 / 1	First try · 2026-07-24
E-d883d9ce56d6c656	Muse Spark 1.2	Rated	32.1	1376	—	9 / 37 / 44	90 / 90	42	37	0 / 1	First try · 2026-08-13
E-444510032bc433fc	Minimax M3	Rated	52.0	1565	—	59 / 21 / 32	112 / 112	55	21	0 / 1	First try · 2026-07-02
E-a0a38f5e77e983bc	Mistral Medium 3.5	Rated	40.9	1460	—	40 / 9 / 47	96 / 116	47	9	20 / 1	Repaired · 2026-07-02
E-fb8801908d28001f	Kimi K3	Rated	46.3	1500	—	32 / 130 / 29	191 / 191	74	130	0 / 1	First try · 2026-07-24
E-0ce5e196701b51f3	Nemotron 3 Ultra 550B A55B	Rated	50.9	1565	—	25 / 13 / 30	68 / 68	34	13	0 / 1	First try · 2026-07-02
E-cb2bd51b4adfa607	O3	Rated	2.8	1083	—	5 / 42 / 111	158 / 158	60	42	0 / 1	First try · 2026-07-21
E-55f6c0e3dc8a8065	GPT-OSS 120B	Rated	17.7	1226	—	27 / 31 / 108	166 / 166	66	31	0 / 1	First try · 2026-07-21
E-8461eb1b16e40129	Laguna S 2.1	Rated	53.4	1581	—	41 / 18 / 41	100 / 100	49	18	0 / 1	Repaired · 2026-07-24
E-c58ec61528547516	Laguna S 2.1	Rated	39.6	1445	—	37 / 32 / 41	110 / 110	52	32	0 / 1	First try · 2026-07-24
E-8a70cd704fcbcf6f	Qwen3.7 Max	Rated	60.4	1652	—	41 / 23 / 16	80 / 86	40	23	0 / 1	First try · 2026-07-02
E-5e09f0c994a59f3c	Qwen3.7 Plus	Rated	27.5	1331	—	18 / 25 / 51	94 / 94	46	25	0 / 1	Repaired · 2026-07-03
E-60f1c4442483d5b0	Qwen3.8 Max	Limited data	25.5	1367	—	2 / 1 / 13	16 / 16	9	1	0 / 1	First try · 2026-08-08
E-5661caf556b12391	Ox Alpha	Rated	43.2	1545	—	8 / 2 / 4	14 / 14	7	2	0 / 1	Repaired · 2026-08-22
E-98157b387f582b30	Ox Alpha	Rated	25.7	1375	—	2 / 0 / 12	14 / 14	7	0	0 / 1	Repaired · 2026-08-22
E-e66ff5a9cbd5132f	Ox Alpha	Rated	27.8	1396	—	2 / 1 / 11	14 / 14	7	1	0 / 1	First try · 2026-08-22
E-3407ff2dd1928eb5	Hy3 Preview	Rated	72.9	1774	—	55 / 14 / 9	78 / 84	39	14	0 / 1	First try · 2026-07-02
E-9e638fc0ac98df4f	Inkling	Rated	19.3	1248	—	15 / 49 / 48	112 / 112	53	49	0 / 1	First try · 2026-07-24
E-bc7c657b08ebac66	Inkling	Rated	5.8	1118	—	6 / 30 / 74	110 / 110	52	30	0 / 1	First try · 2026-07-24
E-52cb66309e46f2f3	Inkling Small	Limited data	39.7	1497	—	9 / 3 / 7	19 / 19	9	3	0 / 1	First try · 2026-08-09
E-7ee1b68d1249ac22	Grok 4.5	Rated	45.8	1495	—	37 / 126 / 34	197 / 197	75	126	0 / 1	First try · 2026-07-24
E-2cdd8ae895784956	MiMo-V2.5	Rated	56.0	1611	—	40 / 6 / 32	78 / 78	39	6	0 / 1	First try · 2026-07-02
E-912a4a40bbdcbd45	MiMo-V2.5	Rated	46.8	1515	—	40 / 25 / 43	108 / 108	53	25	0 / 1	First try · 2026-07-02
E-a168350bd057f924	MiMo-V2.5	Rated	36.3	1416	—	24 / 29 / 39	92 / 92	46	29	0 / 1	First try · 2026-07-01
E-9ce4191c7f1f3b3e	MiMo-V2.5-Pro	Rated	36.9	1419	—	32 / 29 / 47	108 / 108	53	29	0 / 1	First try · 2026-07-02
E-5c3b7f5666a38043	GLM-5.2	Rated	40.9	1465	—	27 / 25 / 26	78 / 78	39	25	0 / 1	First try · 2026-07-11
E-c8076ea61de8da45	GLM-5.2	Rated	16.5	1231	—	22 / 3 / 45	70 / 70	35	3	0 / 1	First try · 2026-07-02