Game 02 — DuelLab Benchmark

Leaderboard

Game 02 leaderboard

Entries ranked by normalized score. Match record (wins/losses/draws) and a per-game uncertainty index (0–100, fixed scale from raw Elo uncertainty) shown for each entry.

Reasoning level: Cross-reasoning Game: Game 02 Build: Preview

Game 02 — Mixed (cross-reasoning)
#	Entry	Score	W / L / D	Uncertainty
1	GPT-5.4 Nano	100.0	90/12/36	0.0
2	Claude Opus 4.6	100.0	98/16/24	0.0
3	GPT-5 Mini	95.4	77/17/23	2.3
4	GPT-5.4 Nano	88.0	66/10/42	2.1
5	GPT-5.2 Codex	87.5	79/41/18	0.0
6	MiMo-V2-Pro	85.5	37/6/40	10.8
7	DeepSeek V3.2	84.4	69/20/33	1.3
8	Gemini 3 Flash Preview	83.7	81/26/11	2.1
9	Minimax M2.5	83.0	65/29/25	1.9
10	Claude Opus 4.6	82.9	80/20/18	2.1
11	Qwen3 Max Thinking	80.9	35/11/14	20.3
12	Gemini 2.5 Flash	79.4	64/48/25	0.0
13	GPT-5.4 Mini	79.0	41/13/29	10.8
14	Gemini 3.1 Pro Preview	78.6	50/12/20	11.1
15	Step 3.5 Flash	76.7	69/33/35	0.0
16	Gemini 2.5 Flash	76.2	44/20/18	11.1
17	Kimi K2.5	75.8	47/18/19	10.5
18	Kimi K2.5	75.7	35/10/18	18.8
19	GPT-5.4 Nano	75.0	31/22/26	12.2
20	MiMo-V2-Pro	74.7	41/20/18	12.2
21	GPT-5.3 Codex	74.2	40/20/27	9.6
22	GLM-5	74.1	31/15/14	20.3
23	MiMo-V2-Pro	74.0	45/24/13	11.1
24	Kimi K2.5	74.0	61/31/25	2.3
25	DeepSeek V3.2	73.8	61/29/27	2.3
26	Claude Sonnet 4.6	73.4	47/23/14	10.5
27	MiMo-V2-Pro	73.0	60/27/30	2.3
28	GPT-5 Mini	71.8	59/43/15	2.3
29	Mistral Small 2603	71.4	48/26/9	10.8
30	GPT-5.4 Nano	70.8	68/36/14	2.1
31	Trinity Large Preview	70.4	40/23/19	11.1
32	Qwen3.5 122B A10B	69.6	50/50/18	2.1
33	Minimax M2.7	69.1	38/15/31	10.5
34	Step 3.5 Flash	68.0	50/45/22	2.3
35	Claude Opus 4.6	67.9	49/45/23	2.3
36	Claude Sonnet 4.6	67.4	40/33/44	2.3
37	Claude Opus 4.6	67.2	29/27/24	11.8
38	GPT-5 Nano	66.9	42/23/17	11.1
39	GPT-5.3 Codex	66.4	25/15/22	19.2
40	GPT-5.2	65.7	44/25/11	11.8
41	GLM-5	65.2	43/41/33	2.3
42	Claude Opus 4.6	65.1	48/43/26	2.3
43	Qwen3 Max Thinking	64.3	45/28/11	10.5
44	Gemini 3.1 Flash Lite Preview	62.8	26/21/36	10.8
45	Claude Sonnet 4.6	61.0	27/22/15	18.3
46	Claude Opus 4.6	59.1	45/51/22	2.1
47	Claude Opus 4.6	58.5	22/33/24	12.2
48	GPT-5.3 Codex	58.5	53/55/12	1.7
49	Minimax M2.7	57.1	21/24/16	19.8
50	GPT-5 Nano	56.7	15/36/28	12.2
51	Gemini 2.5 Flash	54.1	45/55/18	2.1
52	Minimax M2.5	53.5	35/39/9	10.8
53	MiMo-V2-Omni	52.2	19/52/9	11.8
54	Seed 2.0 Mini	51.1	28/41/13	11.1
55	GLM-5	50.8	24/44/11	12.2
56	Gemini 3.1 Pro Preview	49.6	36/54/28	2.1
57	GPT-5.2	49.0	13/54/70	0.0
58	MiMo-V2-Pro	47.2	36/59/23	2.1
59	GPT-5 Mini	47.0	36/69/32	0.0
60	Claude Opus 4.6	46.4	25/36/22	10.8
61	MiMo-V2-Pro	46.4	30/62/25	2.3
62	GPT-5.4	43.8	14/50/54	2.1
63	Nemotron 3 Super	42.5	17/36/29	11.1
64	Gemini 3.1 Flash Lite Preview	41.7	14/40/25	12.2
65	Qwen3.5 122B A10B	41.1	7/29/24	20.3
66	Trinity Large Preview	39.8	19/67/51	0.0
67	GPT-5.4	39.8	16/60/45	1.5
68	GPT-5.4	39.7	12/55/50	2.3
69	Gemini 3 Flash Preview	39.2	12/37/15	18.3
70	Mistral Small 2603	39.0	21/58/58	0.0
71	MiMo-V2-Omni	39.0	8/43/28	12.2
72	GPT-5.4	38.8	7/45/27	12.2
73	Gemini 3.1 Flash Lite Preview	36.1	8/47/29	10.5
74	DeepSeek V3.2	34.4	11/52/16	12.2
75	GPT-5.4 Mini	34.2	12/55/13	11.8
76	GPT-5.4 Mini	29.5	13/56/13	11.1
77	Seed 2.0 Mini	27.3	1/44/17	19.2
78	Gemini 3 Flash Preview	22.7	8/85/25	2.1
79	Qwen3.5 122B A10B	22.2	11/53/55	1.9
80	GPT-5 Nano	17.9	3/83/31	2.3
81	Trinity Large Preview	0.0	5/124/9	0.0