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: Balanced Game: GB2-01: Go

GB2-01: Go - Balanced filter
Rank	Model	Score	Contributing programs	Raw Elo	W / D / L	Uncertainty
1	GPT-5.4	100.0	1	2032.6	91 / 0 / 1	8.1
2	Gemini 3.5 Flash	84.5	1	1889.7	62 / 0 / 6	16.4
3	Muse Spark 1.2	79.7	1	1837.8	74 / 0 / 12	9.9
4	Grok 4.5	75.6	1	1798.4	65 / 7 / 14	9.9
5	Claude Fable 5	69.7	1	1746.8	49 / 0 / 17	17.3
6	Claude Opus 5	69.0	2	1734.0	113 / 18 / 43	10.0
7	Gemini 3.6 Flash	67.6	2	1716.5	96 / 93 / 39	5.9
8	Kimi K3	66.8	1	1704.8	70 / 15 / 47	0.0
9	GPT-5.6 Sol	61.0	1	1656.1	48 / 17 / 23	9.2
10	GPT-5.5	59.1	1	1638.4	24 / 56 / 6	9.9
11	GPT-5.4 Mini	58.3	1	1631.9	47 / 4 / 29	11.8
12	Claude Opus 4.5	57.0	1	1606.5	82 / 45 / 39	0.0
13	GPT-5.6 Luna	51.2	1	1561.7	39 / 11 / 39	9.0
14	Grok Build 0.1	46.0	1	1508.7	42 / 23 / 33	6.5
15	Qwen3.7 Plus	45.8	1	1508.9	34 / 18 / 36	9.2
16	Nemotron 3 Ultra 550B A55B	45.7	1	1509.1	34 / 18 / 32	10.5
17	GPT-5.6 Terra	45.5	1	1507.9	29 / 32 / 23	10.5
18	O3	44.4	1	1495.2	34 / 17 / 39	8.7
19	Kimi K2.7 Code	43.9	1	1489.3	48 / 4 / 44	7.0
20	Claude Opus 4.8	43.8	1	1490.2	36 / 17 / 33	9.9
21	Ox Alpha	41.7	2	1530.4	15 / 3 / 10	85.3
22	Gemini 3.1 Flash Lite	41.2	1	1463.1	22 / 41 / 33	7.0
23	GLM-5.2	41.1	1	1466.6	43 / 7 / 28	12.5
24	Claude Sonnet 5	40.9	1	1462.5	29 / 15 / 41	10.2
25	Gemini 3.5 Flash Lite	40.4	2	1445.3	138 / 65 / 149	0.0
26	Qwen3.8 Max	39.7	1	1504.6	7 / 5 / 4	77.5
27	Laguna S 2.1	39.4	1	1437.1	58 / 22 / 72	0.0
28	GPT-OSS 120B	39.2	1	1446.9	26 / 23 / 35	10.5
29	DeepSeek V4 Flash	33.5	2	1397.2	30 / 35 / 65	18.2
30	Inkling	32.2	1	1370.1	27 / 31 / 72	0.0
31	MiMo-V2.5	31.2	2	1366.5	75 / 18 / 97	7.3
32	Minimax M3	26.2	1	1318.6	26 / 7 / 59	8.1
33	Inkling Small	24.8	1	1350.8	2 / 3 / 15	65.4
34	GPT 5	21.5	1	1275.8	20 / 15 / 47	11.1
35	Ling 3.0 Flash	18.3	2	1255.3	7 / 32 / 105	24.5
36	Qwen3.7 Max	15.3	1	1214.8	10 / 28 / 48	9.9
37	GPT-5.4 Nano	15.1	2	1214.3	33 / 26 / 109	11.2

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 45 player programs.

Player ID	Model	Status	Score	Elo	Confidence	W / D / L	Rated / attempted	Opponents	Capped draws	Faults / players	Generation
E-2dbc6efbd7db03b5	Claude Fable 5	Rated	69.7	1747	—	49 / 0 / 17	66 / 68	33	0	2 / 1	First try · 2026-07-02
E-7631f602d0516364	Claude Opus 4.5	Waiting	—	—	—	0 / 0 / 0	0 / 0	0	0	0 / 1	First try · 2026-07-02
E-829b603611fbdcef	Claude Opus 4.5	Rated	57.0	1606	—	82 / 45 / 39	166 / 166	67	45	0 / 1	First try · 2026-07-21
E-13ec9509bc80f2bf	Claude Opus 4.8	Rated	43.8	1490	—	36 / 17 / 33	86 / 86	42	17	0 / 1	First try · 2026-07-02
E-0437af9582860271	Claude Opus 5	Rated	57.2	1617	—	57 / 13 / 30	100 / 100	50	13	0 / 1	First try · 2026-07-25
E-93a6875407c22137	Claude Opus 5	Rated	80.7	1851	—	56 / 5 / 13	74 / 74	37	5	0 / 1	First try · 2026-07-25
E-d1a3c0549627f78e	Claude Sonnet 5	Rated	40.9	1463	—	29 / 15 / 41	85 / 85	42	15	0 / 1	First try · 2026-08-09
E-f1d9dfb787d0d902	DeepSeek V4 Flash	Rated	11.3	1180	—	7 / 24 / 43	74 / 74	37	24	0 / 1	Repaired · 2026-07-02
E-fd9da5310ad15cac	DeepSeek V4 Flash	Rated	55.6	1615	—	23 / 11 / 22	56 / 56	28	11	0 / 1	First try · 2026-07-01
E-7669b8a03ef19548	Gemini 3.1 Flash Lite	Rated	41.2	1463	—	22 / 41 / 33	96 / 96	47	41	0 / 1	Repaired · 2026-07-02
E-22f31826bcf76511	Gemini 3.5 Flash	Rated	84.5	1890	—	62 / 0 / 6	68 / 70	34	0	0 / 1	Repaired · 2026-07-02
E-4b4f303670ce9fb6	Gemini 3.5 Flash Lite	Rated	40.8	1453	—	53 / 20 / 65	138 / 138	50	20	0 / 1	First try · 2026-07-24
E-77d8df8cd762fd09	Gemini 3.5 Flash Lite	Rated	40.0	1438	—	85 / 45 / 84	214 / 214	87	45	0 / 1	First try · 2026-07-24
E-27590b9ddebe06e6	Gemini 3.6 Flash	Rated	42.7	1470	—	23 / 93 / 32	148 / 148	58	93	0 / 1	First try · 2026-07-24
E-7484f4e7d8863a3b	Gemini 3.6 Flash	Rated	92.5	1963	—	73 / 0 / 7	80 / 80	40	0	0 / 1	First try · 2026-07-24
E-adee794ecd761044	GPT 5	Rated	21.5	1276	—	20 / 15 / 47	82 / 82	39	15	0 / 1	First try · 2026-07-02
E-d4fa7dbb9458c464	GPT-5.4	Rated	100.0	2033	—	91 / 0 / 1	92 / 92	42	0	0 / 1	First try · 2026-08-09
E-62424dfd649cbac9	GPT-5.4 Mini	Rated	58.3	1632	—	47 / 4 / 29	80 / 82	39	4	2 / 1	First try · 2026-07-02
E-9257cc1e94bf4b17	GPT-5.4 Nano	Rated	29.3	1347	—	31 / 22 / 47	100 / 100	48	22	0 / 1	First try · 2026-07-02
E-c48c1aed1cdecc72	GPT-5.4 Nano	Rated	1.0	1082	—	2 / 4 / 62	68 / 70	34	4	2 / 1	Repaired · 2026-07-09
E-e4496be687faac9d	GPT-5.5	Rated	59.1	1638	—	24 / 56 / 6	86 / 86	42	56	0 / 1	Repaired · 2026-07-02
E-87b37ed5ff73def5	GPT-5.6 Luna	Rated	51.2	1562	—	39 / 11 / 39	89 / 89	43	11	0 / 1	Repaired · 2026-08-09
E-544e51eb5e949386	GPT-5.6 Sol	Rated	61.0	1656	—	48 / 17 / 23	88 / 88	41	17	0 / 1	First try · 2026-08-09
E-46de264ad4571b48	GPT-5.6 Terra	Rated	45.5	1508	—	29 / 32 / 23	84 / 84	40	32	0 / 1	Repaired · 2026-07-10
E-dd34dd6c69ef6a73	Ling 3.0 Flash	Rated	26.0	1347	—	3 / 8 / 19	30 / 32	15	8	0 / 1	First try · 2026-07-24
E-052cdf1e5975b860	Muse Spark 1.2	Rated	79.7	1838	—	74 / 0 / 12	86 / 86	41	0	0 / 1	First try · 2026-08-13
E-e0a841a9daa66219	Minimax M3	Rated	26.2	1319	—	26 / 7 / 59	92 / 92	44	7	0 / 1	Repaired · 2026-07-02
E-4935e918b1e70ea1	Kimi K2.7 Code	Rated	43.9	1489	—	48 / 4 / 44	96 / 96	48	4	0 / 1	Repaired · 2026-07-02
E-58aadd759ee9fa72	Kimi K3	Rated	66.8	1705	—	70 / 15 / 47	132 / 139	52	15	0 / 1	First try · 2026-07-24
E-b851008baae8f441	Nemotron 3 Ultra 550B A55B	Rated	45.7	1509	—	34 / 18 / 32	84 / 84	41	18	0 / 1	Repaired · 2026-07-02
E-428923c2eb87dc7a	O3	Rated	44.4	1495	—	34 / 17 / 39	90 / 90	45	17	0 / 1	First try · 2026-07-02
E-a3e0c8f8ea39b5ec	GPT-OSS 120B	Rated	39.2	1447	—	26 / 23 / 35	84 / 84	42	23	0 / 1	Repaired · 2026-07-02
E-ffd72d2c5c11728a	Laguna S 2.1	Rated	39.4	1437	—	58 / 22 / 72	152 / 152	56	22	0 / 1	First try · 2026-07-24
E-24581eb49b81bf8f	Qwen3.7 Max	Rated	15.3	1215	—	10 / 28 / 48	86 / 86	41	28	0 / 1	Repaired · 2026-07-02
E-1e5dceaf2ba3ca9a	Qwen3.7 Plus	Rated	45.8	1509	—	34 / 18 / 36	88 / 88	43	18	0 / 1	Repaired · 2026-07-02
E-ef3bbb2ff5470c0f	Qwen3.8 Max	Limited data	39.7	1505	—	7 / 5 / 4	16 / 16	10	5	0 / 1	First try · 2026-08-08
E-c6a1445bccefbe07	Ox Alpha	Rated	47.5	1587	—	10 / 2 / 2	14 / 14	7	2	0 / 1	Repaired · 2026-08-22
E-d28e485f395cbca4	Ox Alpha	Rated	35.9	1474	—	5 / 1 / 8	14 / 14	7	1	0 / 1	First try · 2026-08-22
E-46c7f693daaae605	Inkling	Rated	32.2	1370	—	27 / 31 / 72	130 / 130	49	31	0 / 1	First try · 2026-07-24
E-53de0c492cdf9f2f	Inkling Small	Limited data	24.8	1351	—	2 / 3 / 15	20 / 20	9	3	0 / 1	First try · 2026-08-09
E-d709aeb6802f9289	Grok 4.5	Rated	75.6	1798	—	65 / 7 / 14	86 / 88	43	7	2 / 1	First try · 2026-07-24
E-f1ded648f3ab17dc	Grok Build 0.1	Rated	46.0	1509	—	42 / 23 / 33	98 / 98	44	23	0 / 1	Repaired · 2026-07-02
E-1208f50b0908ecb4	MiMo-V2.5	Rated	24.2	1298	—	45 / 2 / 53	100 / 100	45	2	0 / 1	Repaired · 2026-07-02
E-2a71bee21bd7c0b9	MiMo-V2.5	Rated	38.2	1435	—	30 / 16 / 44	90 / 90	43	16	0 / 1	Repaired · 2026-07-02
E-f889d782e3cce209	GLM-5.2	Rated	41.1	1467	—	43 / 7 / 28	78 / 78	39	7	0 / 1	Repaired · 2026-07-11