Mystery Game Puzzles (Epoch AI)

#	Model	Effort	Score	Cost / task	Harness
1	GPT-6 Astra	max	84.0%	–	Epoch AI
2	GPT-6.1 Sol	max	80.0%	–	Epoch AI
3	Opus 5.5	max	71.0%	–	Epoch AI
4	Sonnet 5.5	max	65.0%	–	Epoch AI
5	Gemini 4	high	64.0%	–	Epoch AI
6	Fable 5.1	max	58.0%	–	Epoch AI
7	GPT-5.6 Sol	max	58.0%	–	Epoch AI
8	GPT-6 Sol	max	56.0%	–	Epoch AI
9	GPT-5.5	xhigh	56.0%	–	Epoch AI
10	Fable 5	max	52.0%	–	Epoch AI
11	Gemini 3.8 Flash	high	47.0%	–	Epoch AI
12	DeepSeek V4.1 Flash	max	43.0%	–	Epoch AI
13	DeepSeek V4 Pro 0813	max	43.0%	–	Epoch AI
14	Qwen 3.8 Max	xhigh	38.0%	–	Epoch AI
15	GPT-5.4	xhigh	37.0%	–	Epoch AI
16	Gemini 3.7 Flash	high	37.0%	–	Epoch AI
17	Opus 5	max	37.0%	–	Epoch AI
18	Opus 4.8	max	36.0%	–	Epoch AI
19	GPT-5.6 Terra	max	35.0%	–	Epoch AI
20	Grok 4.6	xhigh	34.0%	–	Epoch AI
21	Gemini 3.1 Pro	high	34.0%	–	Epoch AI
22	DeepSeek V4 Flash 0731	max	34.0%	–	Epoch AI
23	GLM 5.3	max	33.0%	–	Epoch AI
24	Qwen 3.7 Max	max	32.0%	–	Epoch AI
25	Gemini 3.5 Flash	high	32.0%	–	Epoch AI
26	Haiku 5.5	max	30.0%	–	Epoch AI
27	Gemini 3.6 Flash	high	30.0%	–	Epoch AI
28	Grok 4.7	xhigh	29.0%	–	Epoch AI
29	o3	high	29.0%	–	Epoch AI
30	Kimi K3	max	26.0%	–	Epoch AI
31	Gemini 3 Flash	low	26.0%	–	Epoch AI
32	Muse Spark 1.3	max	25.0%	–	Epoch AI
33	GPT-5.2	high	23.0%	–	Epoch AI
34	GPT-5	high	23.0%	–	Epoch AI
35	Qwen 3.6 35B-A3B	none	22.0%	–	Epoch AI
36	Opus 4.5	high	22.0%	–	Epoch AI
37	GPT-5.6 Luna	max	21.0%	–	Epoch AI
38	Opus 4.1	high	21.0%	–	Epoch AI
39	Qwen 3.5 Flash	max	20.0%	–	Epoch AI
40	Nemotron 3 Ultra 550B A55B	max	20.0%	–	Epoch AI
41	Qwen 3.6 Max	none	19.0%	–	Epoch AI
42	GPT-5.1	low	19.0%	–	Epoch AI
43	GLM 5.2	low	19.0%	–	Epoch AI
44	Gemini 3.5 Flash-Lite	low	19.0%	–	Epoch AI
45	Kimi K2.6	max	18.0%	–	Epoch AI
46	Qwen 3.6 Flash	none	18.0%	–	Epoch AI
47	Qwen 3.5 397B-A17B	none	18.0%	–	Epoch AI
48	Mistral Large 4	high	17.0%	–	Epoch AI
49	Qwen 3.7 Plus	high	17.0%	–	Epoch AI
50	Qwen 3.5 122B-A10B	none	17.0%	–	Epoch AI
51	Qwen 3.5 Plus	none	17.0%	–	Epoch AI
52	DeepSeek V4 Pro	none	17.0%	–	Epoch AI
53	Sonnet 4.5	high	17.0%	–	Epoch AI
54	Sonnet 4.6	low	16.0%	–	Epoch AI
55	Sonnet 5	max	16.0%	–	Epoch AI
56	Qwen 3.7 Flash	max	15.0%	–	Epoch AI
57	Opus 4.6	max	15.0%	–	Epoch AI
58	Opus 4.7	max	13.0%	–	Epoch AI
59	GPT-4	high	12.0%	–	Epoch AI
60	Qwen 3.6 Plus	none	12.0%	–	Epoch AI
61	GPT-4o mini	high	12.0%	–	Epoch AI
62	GPT-5.4 Mini	none	11.0%	–	Epoch AI
63	GPT-5 Mini	minimal	10.0%	–	Epoch AI
64	GPT-5.4 Nano	none	9.0%	–	Epoch AI
65	GPT-5 Nano	medium	9.0%	–	Epoch AI
66	Qwen 3 235B-A22B	xhigh	9.0%	–	Epoch AI
67	GLM 5.3 Flash	max	8.0%	–	Epoch AI
68	MiniMax M3	none	8.0%	–	Epoch AI
69	GPT-6 Luna	max	7.0%	–	Epoch AI
70	o3 mini	high	7.0%	–	Epoch AI
71	Qwen 3.6 27B	none	7.0%	–	Epoch AI
72	GPT-4.1 Mini	high	7.0%	–	Epoch AI
73	Inkling Small	xhigh	6.0%	–	Epoch AI
74	o4 mini	high	5.0%	–	Epoch AI
75	Qwen 3 Max	max	5.0%	–	Epoch AI
76	GPT-3.5	high	3.0%	–	Epoch AI
77	gpt-oss-120b	medium	2.0%	–	Epoch AI