Past progressive (was/were + -ing) overuse

Test: Bad Writing Habits

Avg. Score

88.1%

Scenarios

Overall Performance

Performance Score Distribution (Top 20)

Click a model name to view its detail page.

Rank ▲	Model	Score	Avg. Cost	Avg. Time	Stability
1	GPT-4.1 Mini	99.9%	$0.0027	19.0s	99%
2	Inception Mercury 2	99.7%	$0.0032	7.0s	96%
3	Grok 4 Fast	100.0%	$0.0017	24.1s	99%
4	LFM2 24B	99.9%	$0.0002	28.4s	98%
5	Grok 4.1 Fast	100.0%	$0.0018	37.8s	100%
6	o4 Mini	99.8%	$0.015	25.7s	97%
7	GPT-5.4 Nano	99.2%	$0.0057	26.3s	93%
8	GPT-4o, Aug. 6th (temp=1)	99.7%	$0.018	24.4s	95%
9	GPT-5.4 Nano (Reasoning, Low)	99.0%	$0.0055	20.6s	90%
10	GPT-4.1	99.8%	$0.018	44.7s	96%
11	o4 Mini High	100.0%	$0.025	47.2s	100%
12	Stealth: Aurora Alpha	98.1%	$0.0000	9.8s	82%
13	GPT-5 Mini	99.7%	$0.0100	57.4s	96%
14	GPT-5.4 Mini (Reasoning, Low)	98.2%	$0.015	16.8s	88%
15	GPT-5 Nano	99.6%	$0.0042	1.4m	95%
16	GPT-5.4 Mini (Reasoning)	98.6%	$0.022	28.1s	88%
17	GPT-4o Mini (temp=1)	97.7%	$0.0012	34.8s	77%
18	GPT-4o Mini (temp=0)	97.9%	$0.0012	34.8s	76%
19	GPT-5.4 Nano (Reasoning)	96.8%	$0.0061	24.5s	76%
20	Nemotron 3 Nano	98.1%	$0.0010	1.1m	81%
21	Nemotron 3 Super	98.5%	$0.0000	1.4m	83%
22	GPT-5.4	100.0%	$0.049	1.4m	100%
23	Grok 4.20 (Beta)	95.8%	$0.018	15.8s	73%
24	GPT-4o, Aug. 6th (temp=0)	97.0%	$0.023	22.7s	73%
25	Grok 4	100.0%	$0.048	1.7m	100%
26	GPT-5.2	100.0%	$0.056	1.5m	100%
27	GPT-5.4 Mini	96.2%	$0.015	16.8s	68%
28	Llama 3.1 Nemotron 70B	95.4%	$0.0038	31.7s	67%
29	GPT-4o, May 13th (temp=1)	96.9%	$0.033	14.4s	71%
30	GPT-5.1	100.0%	$0.054	1.8m	99%
31	Claude 3 Haiku	92.7%	$0.0025	14.9s	63%
32	Mistral Small 4 (Reasoning)	92.9%	$0.0022	30.2s	66%
33	ByteDance Seed 1.6 Flash	93.9%	$0.0013	27.3s	61%
34	DeepSeek V3 (2025-03-24)	94.5%	$0.0014	39.4s	63%
35	Qwen 3.5 Plus (2026-02-15)	94.4%	$0.0060	31.5s	62%
36	Writer: Palmyra X5	94.3%	$0.011	22.0s	62%
37	Qwen3 235B A22B Instruct 2507	94.8%	$0.0011	59.2s	65%
38	Grok 4.20 (Beta, Reasoning)	96.3%	$0.039	34.0s	71%
39	GPT-5.4 (Reasoning, Low)	99.0%	$0.055	1.4m	87%
40	GPT-4o, May 13th (temp=0)	93.9%	$0.035	14.1s	67%
41	Qwen 2.5 72B	91.6%	$0.0010	36.7s	62%
42	Gemini 2.5 Flash (Reasoning)	92.8%	$0.011	21.5s	57%
43	Hermes 3 405B	93.3%	$0.0032	53.2s	62%
44	Gemini 2.5 Flash Lite	91.3%	$0.0009	9.5s	52%
45	Gemini 2.5 Flash	92.0%	$0.0052	10.6s	53%
46	DeepSeek-V2 Chat	92.9%	$0.0021	53.3s	61%
47	Arcee AI: Trinity Mini	90.5%	$0.0003	9.2s	51%
48	DeepSeek V3 (2024-12-26)	92.6%	$0.0021	54.6s	58%
49	Hermes 3 70B	93.9%	$0.0010	1.2m	58%
50	Z.AI GLM 4.5	91.5%	$0.0051	42.1s	56%
51	Mistral Small 4	90.1%	$0.0014	18.2s	49%
52	GPT-5	100.0%	$0.065	2.8m	100%
53	Claude 3.7 Sonnet	95.0%	$0.042	46.7s	64%
54	Cohere Command R+ (Aug. 2024)	93.1%	$0.020	52.5s	58%
55	Claude Sonnet 4.5	92.9%	$0.035	38.1s	60%
56	Gemini 2.5 Flash Lite (Reasoning)	89.7%	$0.0028	30.8s	49%
57	Mistral Medium 3.1	88.6%	$0.0048	36.5s	50%
58	Mistral NeMO	86.3%	$0.0005	10.1s	45%
59	Llama 3.1 8B	92.1%	$0.0003	1.3m	51%
60	Claude 3.5 Sonnet	92.6%	$0.048	35.5s	59%
61	GPT-5.4 (Reasoning)	99.4%	$0.089	2.6m	93%
62	Qwen 3 32B	89.1%	$0.0015	54.6s	45%
63	GPT-4.1 Nano	84.8%	$0.0007	13.3s	39%
64	Claude Sonnet 4	89.6%	$0.032	43.7s	52%
65	Mistral Large 3	85.0%	$0.0033	30.3s	42%
66	Mistral Large 2	85.7%	$0.013	29.4s	45%
67	MoonshotAI: Kimi K2.5	95.7%	$0.019	3.2m	73%
68	ByteDance Seed 2.0 Lite	92.8%	$0.012	2.2m	59%
69	Arcee AI: Trinity Large (Preview)	85.7%	$0.0000	43.6s	43%
70	ByteDance Seed 1.6	94.0%	$0.013	2.5m	60%
71	Gemma 3 4B	82.4%	$0.0002	20.0s	41%
72	Llama 3.1 70B	84.5%	$0.0015	29.4s	38%
73	Qwen 3.5 9B	88.8%	$0.0011	1.4m	43%
74	Qwen 3.5 Flash	85.9%	$0.0025	47.5s	39%
75	Rocinante 12B	84.6%	$0.0014	38.4s	38%
76	WizardLM 2 8x22b	89.1%	$0.0026	1.8m	48%
77	Gemma 3 12B	83.0%	$0.0004	41.3s	36%
78	Mistral Large	83.3%	$0.014	30.9s	37%
79	Mistral Small Creative	80.5%	$0.0007	9.1s	31%
80	Aion 2.0	84.8%	$0.0064	1.3m	43%
81	Inception Mercury	83.3%	$0.011	17.6s	29%
82	MiniMax M2.7	83.8%	$0.0040	1.1m	38%
83	Qwen 3.5 397B A17B	91.3%	$0.014	3.0m	56%
84	Z.AI GLM 5 Turbo	79.1%	$0.0081	33.2s	34%
85	Gemini 3.1 Flash Lite (Preview)	75.4%	$0.0030	8.4s	31%
86	Ministral 3 3B	77.7%	$0.0005	11.1s	24%
87	Qwen 3.5 122B	85.5%	$0.025	1.1m	34%
88	Stealth: Hunter Alpha	78.2%	$0.0000	55.0s	34%
89	Stealth: Healer Alpha	75.7%	$0.0000	23.7s	30%
90	DeepSeek V3.2	81.5%	$0.0014	1.9m	39%
91	Z.AI GLM 5	81.3%	$0.0084	1.2m	31%
92	Claude Opus 4.5	84.5%	$0.070	53.4s	45%
93	Gemini 3.1 Pro (Preview)	92.7%	$0.107	1.8m	57%
94	Qwen 3.5 35B	78.9%	$0.018	1.0m	31%
95	Claude Haiku 4.5	73.6%	$0.011	21.6s	26%
96	Claude 3.5 Haiku	73.4%	$0.0035	10.8s	19%
97	Claude Opus 4.6	85.4%	$0.078	1.2m	44%
98	Ministral 8B	69.4%	$0.0004	10.4s	23%
99	Gemma 3 27B	74.1%	$0.0006	52.6s	25%
100	Gemini 2.5 Pro	75.2%	$0.036	36.2s	34%
101	Gemini 3 Flash (Preview, Reasoning)	72.7%	$0.012	30.1s	26%
102	Ministral 3 14B	69.3%	$0.0007	11.7s	21%
103	ByteDance Seed 2.0 Mini	90.9%	$0.0045	4.9m	56%
104	MiniMax M2.5	73.6%	$0.0034	1.3m	27%
105	Qwen 3.5 27B	80.2%	$0.020	1.6m	26%
106	Ministral 3 8B	68.2%	$0.0008	19.6s	18%
107	Ministral 3B	66.8%	$0.0001	8.1s	16%
108	Claude Opus 4.6 (Reasoning)	84.8%	$0.088	1.4m	39%
109	Gemini 3 Flash (Preview)	63.0%	$0.0078	19.6s	16%
110	DeepSeek V3.1	69.5%	$0.0020	1.8m	24%
111	Gemini 3 Pro (Preview)	71.9%	$0.055	54.4s	28%
112	Claude Sonnet 4.6	67.1%	$0.031	39.3s	18%
113	Claude Opus 4	91.7%	$0.209	1.4m	55%
114	Z.AI GLM 4.6	61.6%	$0.0065	51.5s	14%
115	Claude Sonnet 4.6 (Reasoning)	69.4%	$0.060	1.2m	20%
116	Z.AI GLM 4.7	60.9%	$0.010	1.4m	12%
117	Z.AI GLM 4.7 Flash	51.6%	$0.0017	1.2m	8%
118	Mistral Small 3.2 24B	80.5%	$0.0069	5.7m	30%
88.08%

Individual Scenarios

Detailed Writing Rules

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
MiniMax M2.5	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
Claude Haiku 4.5	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	98	99.7%
Qwen 3.5 35B	100	100	100	100	98	99.6%
Mistral Small 4	100	100	100	100	98	99.6%
GPT-4o, May 13th (temp=1)	100	100	100	100	97	99.4%
MiniMax M2.7	100	100	100	100	96	99.3%
DeepSeek V3.1	100	100	100	100	94	98.8%
Mistral Medium 3.1	100	100	100	100	94	98.8%
Qwen 3.5 9B	100	100	100	100	93	98.7%
Inception Mercury	100	100	100	100	93	98.6%
Stealth: Healer Alpha	100	100	100	100	90	97.9%
Aion 2.0	100	100	100	100	89	97.8%
ByteDance Seed 1.6 Flash	100	100	100	100	82	96.4%
GPT-4.1 Nano	100	100	100	100	78	95.6%
Claude Sonnet 4	100	100	100	100	73	94.5%
Cohere Command R+ (Aug. 2024)	100	100	100	100	71	94.2%
Claude Sonnet 4.6	100	100	100	100	71	94.2%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	70	94.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	70	94.0%
Mistral Large 2	100	100	100	100	70	94.0%
Qwen 3.5 27B	100	100	100	100	69	93.7%
Stealth: Hunter Alpha	100	100	100	100	66	93.2%
Gemma 3 27B	100	100	100	86	77	92.7%
Claude Sonnet 4.5	100	100	100	82	79	92.2%
Mistral Large 3	100	100	100	79	78	91.3%
Mistral Small Creative	100	100	100	80	75	91.0%
DeepSeek V3.2	100	100	100	100	50	90.0%
Z.AI GLM 4.6	100	100	100	100	48	89.6%
Qwen 2.5 72B	100	100	100	74	73	89.4%
Qwen 3.5 Flash	100	100	100	100	45	89.0%
Gemma 3 4B	100	100	100	100	45	89.0%
Hermes 3 405B	100	100	100	100	33	86.7%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	27	85.5%
WizardLM 2 8x22b	100	100	100	100	12	82.4%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	6	81.2%
Claude 3.5 Haiku	100	100	100	100	0	80.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	0	80.0%
Arcee AI: Trinity Mini	100	100	100	100	0	80.0%
Llama 3.1 8B	100	100	100	100	0	80.0%
Gemini 3 Pro (Preview)	100	100	100	89	0	77.8%
Mistral Large	100	100	97	67	23	77.4%
Z.AI GLM 4.7	100	100	100	44	31	74.9%
Mistral Small 3.2 24B	100	100	86	44	42	74.5%
Qwen 3 32B	100	100	100	68	0	73.6%
Ministral 3B	100	100	100	50	0	70.0%
Ministral 3 8B	100	100	100	28	8	67.1%
Ministral 3 14B	100	100	100	2	0	60.4%
Z.AI GLM 4.7 Flash	100	100	29	5	0	46.8%
Gemini 3 Flash (Preview)	93	84	21	20	0	43.5%
Ministral 3 3B	100	89	0	0	0	37.8%
Ministral 8B	63	52	50	5	0	34.0%

▼

Horror: alone in an eerie place at night

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	99.9%
Claude Opus 4.6 (Reasoning)	100	100	100	100	98	99.6%
Z.AI GLM 5	100	100	100	100	98	99.6%
ByteDance Seed 2.0 Lite	100	100	100	100	95	99.1%
DeepSeek V3 (2024-12-26)	100	100	100	100	93	98.6%
Claude Opus 4	100	100	100	100	91	98.1%
Cohere Command R+ (Aug. 2024)	100	100	100	100	88	97.6%
MiniMax M2.7	100	100	100	100	86	97.3%
Claude Opus 4.6	100	100	100	100	85	96.9%
ByteDance Seed 1.6	100	100	100	100	82	96.5%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	93	89	96.4%
Mistral Large 3	100	100	100	93	89	96.4%
Claude Opus 4.5	100	100	100	95	83	95.7%
Claude 3 Haiku	100	100	100	100	78	95.6%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	77	95.5%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	77	95.3%
Inception Mercury	100	100	100	100	76	95.2%
GPT-4o Mini (temp=1)	100	100	100	100	72	94.5%
DeepSeek V3 (2025-03-24)	100	100	100	100	69	93.9%
GPT-5 Nano	100	100	99	89	80	93.5%
Ministral 3 3B	100	100	100	94	67	92.2%
Llama 3.1 Nemotron 70B	100	100	100	85	75	92.1%
ByteDance Seed 2.0 Mini	100	100	100	100	60	91.9%
Mistral Small 3.2 24B	100	100	100	100	59	91.8%
Qwen 3.5 35B	100	100	100	96	58	90.8%
Z.AI GLM 4.5	100	100	100	80	69	89.9%
Ministral 3B	100	100	100	89	48	87.5%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	37	87.3%
GPT-4o, May 13th (temp=0)	100	100	100	87	46	86.7%
Claude Sonnet 4.6	100	100	100	100	25	84.9%
Mistral Small Creative	100	100	100	100	20	84.0%
Stealth: Aurora Alpha	100	100	100	72	46	83.5%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	17	83.5%
Z.AI GLM 5 Turbo	100	100	100	91	25	83.1%
Aion 2.0	100	100	97	61	53	82.1%
Gemini 2.5 Pro	100	100	92	59	49	80.0%
Claude Sonnet 4.5	100	100	100	100	0	80.0%
Mistral Large	100	100	100	100	0	80.0%
Ministral 3 14B	100	100	100	100	0	80.0%
GPT-4.1 Nano	100	100	100	64	34	79.6%
Arcee AI: Trinity Mini	100	100	100	55	41	79.3%
Claude 3.7 Sonnet	100	100	100	95	0	79.0%
Gemini 3 Pro (Preview)	100	100	100	90	1	78.3%
Llama 3.1 70B	100	100	86	54	48	77.7%
Mistral Large 2	100	100	100	88	0	77.6%
Arcee AI: Trinity Large (Preview)	100	100	100	87	0	77.3%
Gemma 3 4B	100	100	92	56	23	74.2%
WizardLM 2 8x22b	100	100	100	48	21	73.8%
Claude 3.5 Haiku	100	100	100	61	0	72.2%
Qwen 2.5 72B	100	100	69	66	25	72.2%
DeepSeek V3.2	100	100	97	48	0	68.9%
Hermes 3 405B	100	100	100	37	0	67.5%
Rocinante 12B	100	100	100	28	6	66.9%
MiniMax M2.5	100	100	64	34	32	66.1%
Writer: Palmyra X5	100	100	80	41	0	64.3%
Claude Sonnet 4	100	100	72	46	0	63.6%
Stealth: Healer Alpha	100	100	48	46	16	62.1%
Gemini 3 Flash (Preview)	100	100	69	33	0	60.4%
Z.AI GLM 4.7	100	100	100	0	0	60.0%
Llama 3.1 8B	100	100	100	0	0	60.0%
Gemini 2.5 Flash Lite	100	100	80	0	0	56.0%
Mistral NeMO	100	66	64	26	21	55.3%
Z.AI GLM 4.6	100	74	66	30	0	54.1%
Ministral 8B	100	80	71	0	0	50.1%
Gemma 3 27B	73	73	69	14	0	45.8%
DeepSeek V3.1	100	68	33	0	0	40.2%
Claude Haiku 4.5	100	88	10	1	0	39.5%
Stealth: Hunter Alpha	100	81	10	0	0	38.4%
Gemini 3.1 Flash Lite (Preview)	80	46	23	22	14	37.2%
Ministral 3 8B	100	40	28	4	0	34.4%
Z.AI GLM 4.7 Flash	100	37	29	0	0	33.3%

▼

Literary fiction: old friends reunite

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
MiniMax M2.5	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
GPT-4.1 Nano	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	98	99.6%
Aion 2.0	100	100	100	100	95	98.9%
Z.AI GLM 4.5	100	100	100	97	96	98.5%
Hermes 3 405B	100	100	100	100	92	98.5%
Qwen 3.5 35B	100	100	100	100	90	98.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	89	97.8%
Mistral Medium 3.1	100	100	100	100	87	97.4%
Ministral 3B	100	100	100	100	84	96.8%
Gemini 2.5 Flash	100	100	100	100	80	96.0%
Mistral Small 4	100	100	100	100	77	95.3%
Ministral 3 14B	100	100	100	100	67	93.5%
Gemma 3 12B	100	100	100	97	65	92.3%
Qwen 3.5 Flash	100	100	100	100	61	92.1%
Z.AI GLM 5	100	100	100	100	53	90.7%
Stealth: Hunter Alpha	100	100	100	86	65	90.1%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	44	88.8%
DeepSeek V3.2	100	100	100	85	56	88.2%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	96	45	88.2%
Nemotron 3 Super	100	100	100	97	40	87.4%
Mistral Small Creative	100	100	100	100	36	87.2%
Mistral Small 4 (Reasoning)	100	100	100	100	36	87.2%
Gemma 3 27B	100	100	84	75	73	86.3%
Arcee AI: Trinity Large (Preview)	100	100	94	85	46	85.1%
Qwen3 235B A22B Instruct 2507	100	100	100	100	21	84.3%
Claude 3.7 Sonnet	100	100	100	95	25	84.0%
MiniMax M2.7	100	100	100	100	15	83.0%
Claude Sonnet 4.5	100	100	92	77	45	82.9%
Gemini 2.5 Pro	100	100	96	80	36	82.4%
Stealth: Healer Alpha	100	100	100	100	4	80.8%
Llama 3.1 70B	100	100	100	100	0	80.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	0	80.0%
Arcee AI: Trinity Mini	100	100	100	93	0	78.7%
Gemini 3 Flash (Preview)	100	100	89	69	0	71.5%
Ministral 3 8B	100	100	72	50	28	70.2%
Claude Haiku 4.5	100	100	100	49	0	69.9%
Z.AI GLM 5 Turbo	100	87	86	75	0	69.6%
Gemini 3 Pro (Preview)	100	100	84	57	0	68.2%
Ministral 8B	100	100	91	32	0	64.7%
Claude Sonnet 4.6 (Reasoning)	100	100	80	29	0	61.7%
DeepSeek V3.1	100	100	70	15	0	56.9%
Rocinante 12B	100	66	57	22	0	49.1%
Z.AI GLM 4.7	95	82	47	20	0	48.7%
Z.AI GLM 4.7 Flash	100	71	48	0	0	43.7%

▼

Mystery: examining a crime scene

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	96	99.3%
Claude 3 Haiku	100	100	100	100	94	98.8%
Llama 3.1 70B	100	100	100	100	92	98.5%
Cohere Command R+ (Aug. 2024)	100	100	100	100	92	98.5%
Mistral Large	100	100	100	100	82	96.3%
Qwen 3 32B	100	100	100	96	84	96.1%
Claude 3.7 Sonnet	100	100	100	100	80	96.0%
ByteDance Seed 1.6	100	100	100	100	68	93.7%
Mistral Large 3	100	100	100	86	80	93.0%
Gemini 2.5 Flash Lite	100	100	100	100	64	92.8%
Llama 3.1 8B	100	100	100	100	58	91.6%
Z.AI GLM 4.6	100	100	100	82	76	91.5%
Gemma 3 4B	100	100	100	80	76	91.1%
Qwen 3.5 Flash	100	100	100	75	74	89.8%
Qwen 3.5 35B	100	100	100	100	49	89.8%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	47	89.4%
Claude Opus 4	100	100	100	100	34	86.8%
Mistral Large 2	100	100	91	72	71	86.8%
Claude Sonnet 4.5	100	100	100	100	31	86.1%
Mistral NeMO	100	100	100	100	29	85.8%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	28	85.6%
Gemini 2.5 Flash	100	100	100	100	23	84.6%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	81	41	84.6%
Qwen 3.5 27B	100	100	100	100	20	84.0%
DeepSeek V3.2	100	100	100	100	19	83.7%
Qwen 2.5 72B	100	100	100	79	37	83.2%
Stealth: Hunter Alpha	100	100	100	100	15	83.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	15	83.0%
Aion 2.0	100	100	100	100	14	82.7%
Gemma 3 12B	100	100	100	100	8	81.5%
Mistral Small 4 (Reasoning)	100	100	90	81	36	81.4%
MiniMax M2.7	100	100	100	100	7	81.4%
ByteDance Seed 2.0 Lite	100	100	100	100	4	80.8%
Inception Mercury	100	100	100	100	0	80.0%
ByteDance Seed 1.6 Flash	100	100	100	100	0	80.0%
Mistral Small Creative	100	100	100	100	0	80.0%
Ministral 3 3B	100	100	100	100	0	80.0%
DeepSeek V3.1	100	100	100	80	20	79.9%
GPT-4o Mini (temp=0)	100	100	96	94	4	78.8%
DeepSeek V3 (2024-12-26)	100	100	100	91	0	78.1%
Claude Haiku 4.5	100	100	100	67	17	76.7%
GPT-4o, May 13th (temp=0)	100	100	99	43	41	76.7%
Mistral Small 4	100	100	100	79	0	75.8%
Arcee AI: Trinity Mini	100	100	92	71	0	72.6%
Z.AI GLM 5 Turbo	100	100	100	60	0	72.0%
Mistral Small 3.2 24B	100	100	100	53	0	70.6%
MiniMax M2.5	100	100	100	40	0	68.1%
Claude Opus 4.5	100	100	86	48	5	68.0%
Ministral 3B	100	100	100	39	0	67.9%
Gemini 3.1 Flash Lite (Preview)	100	100	100	15	15	65.9%
Qwen3 235B A22B Instruct 2507	100	100	92	29	0	64.3%
Claude 3.5 Haiku	100	100	100	15	0	63.0%
Claude Sonnet 4.6	100	94	54	51	11	62.1%
Arcee AI: Trinity Large (Preview)	100	100	37	36	28	60.1%
Z.AI GLM 5	100	100	100	0	0	60.0%
Stealth: Healer Alpha	100	100	100	0	0	60.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	0	0	60.0%
Gemma 3 27B	100	100	100	0	0	60.0%
Gemini 3 Pro (Preview)	100	100	56	24	0	55.9%
GPT-4.1 Nano	100	100	73	5	0	55.8%
Ministral 3 14B	100	88	64	0	0	50.4%
Gemini 2.5 Pro	100	100	26	11	0	47.4%
Gemini 3 Flash (Preview)	100	68	46	19	0	46.6%
Ministral 8B	100	83	33	0	0	43.3%
Ministral 3 8B	100	62	29	0	0	38.2%
Z.AI GLM 4.7	100	47	0	0	0	29.4%
Z.AI GLM 4.7 Flash	62	21	0	0	0	16.7%

▼

Romance: separated couple reunites

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Qwen 3.5 35B	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
DeepSeek V3.1	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	99	99.8%
Nemotron 3 Super	100	100	100	100	97	99.3%
Z.AI GLM 4.7	100	100	100	100	96	99.3%
Claude Sonnet 4.5	100	100	100	100	93	98.7%
Qwen 3.5 Flash	100	100	100	100	88	97.6%
Z.AI GLM 4.5	100	100	100	100	87	97.4%
Cohere Command R+ (Aug. 2024)	100	100	100	100	84	96.8%
DeepSeek V3.2	100	100	100	100	84	96.7%
Claude Opus 4.5	100	100	100	96	80	95.2%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	75	95.0%
Claude Haiku 4.5	100	100	100	90	84	94.8%
Z.AI GLM 5	100	100	100	100	73	94.7%
Arcee AI: Trinity Mini	100	100	100	100	71	94.1%
Gemma 3 27B	100	100	100	100	70	94.0%
Hermes 3 405B	100	100	100	100	67	93.3%
Gemini 3.1 Flash Lite (Preview)	100	100	100	84	80	92.9%
Stealth: Hunter Alpha	100	100	100	100	59	91.8%
Claude 3 Haiku	100	100	100	100	57	91.4%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	54	90.9%
Gemma 3 4B	100	100	100	75	75	90.1%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	50	90.0%
Gemini 3 Pro (Preview)	100	100	100	90	58	89.4%
Ministral 8B	100	100	100	91	53	88.9%
ByteDance Seed 1.6 Flash	100	100	100	93	48	88.1%
Gemini 2.5 Flash Lite	100	100	100	100	34	86.9%
GPT-4.1 Nano	100	100	100	100	30	86.0%
Mistral Large 3	100	100	100	91	38	85.9%
Gemini 3 Flash (Preview)	100	100	100	100	29	85.8%
Mistral Medium 3.1	100	100	100	98	31	85.8%
Arcee AI: Trinity Large (Preview)	100	100	100	71	50	84.1%
Ministral 3 14B	100	100	100	100	15	83.0%
Aion 2.0	100	100	100	83	30	82.6%
Mistral NeMO	100	100	100	93	15	81.6%
Ministral 3 8B	100	88	84	73	62	81.4%
Gemini 2.5 Pro	100	100	100	64	40	80.8%
Claude Sonnet 4.6	100	100	100	100	0	80.0%
Stealth: Healer Alpha	100	100	100	100	0	80.0%
Mistral Small 4	100	100	100	100	0	80.0%
Llama 3.1 8B	100	100	100	100	0	80.0%
MiniMax M2.5	100	100	100	65	26	78.2%
Gemma 3 12B	100	100	100	86	4	78.0%
Ministral 3B	100	100	85	79	0	72.8%
Z.AI GLM 4.7 Flash	100	100	87	53	11	70.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	86	31	26	68.7%
Z.AI GLM 5 Turbo	100	100	100	31	6	67.3%
MiniMax M2.7	100	100	100	16	0	63.1%
Mistral Small 3.2 24B	100	100	100	6	0	61.2%
Rocinante 12B	100	100	100	4	0	60.8%
Z.AI GLM 4.6	100	100	100	0	0	60.0%
Llama 3.1 70B	100	100	100	0	0	60.0%

▼

Thriller: chase through city streets

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Qwen 3.5 35B	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	98	99.6%
MiniMax M2.5	100	100	100	100	92	98.3%
ByteDance Seed 1.6 Flash	100	100	100	100	88	97.7%
Claude Sonnet 4.6	100	100	100	100	87	97.4%
Rocinante 12B	100	100	100	100	87	97.4%
ByteDance Seed 2.0 Mini	100	100	100	100	85	97.0%
Ministral 8B	100	100	100	100	84	96.8%
Claude Opus 4.6 (Reasoning)	100	100	100	94	84	95.5%
Claude Opus 4.6	100	100	100	100	77	95.3%
Mistral Medium 3.1	100	100	100	97	80	95.2%
Qwen 3.5 Flash	100	100	100	100	70	93.9%
WizardLM 2 8x22b	100	100	100	100	65	93.1%
Z.AI GLM 4.7 Flash	100	100	100	100	49	89.9%
Arcee AI: Trinity Large (Preview)	100	100	100	100	46	89.2%
Qwen 3 32B	100	100	100	100	46	89.2%
Llama 3.1 Nemotron 70B	100	100	100	100	45	89.0%
Claude 3.5 Sonnet	100	100	100	100	43	88.6%
Aion 2.0	100	100	100	100	41	88.3%
DeepSeek V3.2	100	100	100	84	55	87.9%
Mistral Large 3	100	100	100	100	38	87.6%
DeepSeek V3 (2025-03-24)	100	100	100	100	37	87.5%
ByteDance Seed 2.0 Lite	100	100	100	100	36	87.3%
Claude Opus 4.5	100	100	100	95	40	87.1%
Mistral Small 4 (Reasoning)	100	100	100	97	32	86.0%
GPT-4o, May 13th (temp=0)	100	100	85	68	68	84.3%
Z.AI GLM 5 Turbo	100	100	100	100	21	84.2%
GPT-4o Mini (temp=1)	100	100	100	100	19	83.8%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	11	82.2%
Qwen 2.5 72B	100	100	100	57	53	81.8%
Mistral NeMO	100	100	92	67	41	80.1%
Qwen 3.5 27B	100	100	100	100	0	80.0%
MiniMax M2.7	100	100	100	100	0	80.0%
Claude Sonnet 4.5	100	100	100	100	0	80.0%
DeepSeek-V2 Chat	100	100	100	100	0	80.0%
Gemini 2.5 Flash	100	100	100	100	0	80.0%
Mistral Small Creative	100	100	100	100	0	80.0%
Arcee AI: Trinity Mini	100	100	100	100	0	80.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	70	26	79.3%
Claude Sonnet 4	100	100	100	95	0	78.9%
Hermes 3 70B	100	100	100	68	25	78.5%
Z.AI GLM 5	100	100	100	64	25	77.7%
GPT-4.1 Nano	100	100	95	93	0	77.7%
Gemini 3 Flash (Preview)	100	100	99	81	8	77.6%
GPT-4o, Aug. 6th (temp=0)	100	100	100	77	10	77.2%
Claude Haiku 4.5	100	100	100	52	33	77.0%
Stealth: Healer Alpha	100	95	87	84	0	73.3%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	60	0	71.9%
Mistral Large 2	100	100	93	47	19	71.8%
Claude Sonnet 4.6 (Reasoning)	100	100	94	63	0	71.3%
Stealth: Hunter Alpha	100	100	65	51	35	70.3%
Ministral 3 8B	100	100	100	37	0	67.5%
Llama 3.1 70B	100	100	100	35	0	66.9%
DeepSeek V3.1	100	100	58	51	21	66.0%
Claude 3.5 Haiku	100	87	74	61	0	64.5%
Gemini 3 Pro (Preview)	100	77	59	42	26	60.9%
Mistral Small 4	100	100	71	28	0	59.9%
Ministral 3 14B	100	100	47	46	0	58.6%
Gemini 2.5 Pro	100	53	42	38	32	53.1%
Gemini 3.1 Flash Lite (Preview)	100	100	26	17	0	48.6%
Gemma 3 12B	100	65	48	8	0	44.2%
Gemma 3 4B	100	94	27	0	0	44.2%
Ministral 3 3B	100	100	15	0	0	43.0%
Mistral Small 3.2 24B	100	100	0	0	0	40.0%
Ministral 3B	100	100	0	0	0	40.0%
Mistral Large	100	69	10	0	0	35.8%
Z.AI GLM 4.6	100	45	25	0	0	34.1%
Gemma 3 27B	100	57	11	0	0	33.6%
Z.AI GLM 4.7	89	10	8	5	0	22.5%

genre

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Claude Opus 4.6 (Reasoning)	100	100	100	100	97	99.4%
DeepSeek V3.2	100	100	100	100	93	98.6%
LFM2 24B	100	100	100	100	91	98.3%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	96	94	98.1%
Claude Opus 4.6	100	100	100	100	90	98.0%
Qwen 3 32B	100	100	100	100	86	97.3%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	85	97.0%
GPT-5.4 Nano	100	100	100	100	85	97.0%
Hermes 3 70B	100	100	100	100	84	96.7%
Llama 3.1 8B	100	100	100	100	79	95.8%
GPT-4o, May 13th (temp=1)	100	100	100	100	78	95.6%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	94	84	95.6%
ByteDance Seed 1.6 Flash	100	100	100	100	77	95.4%
GPT-4.1 Nano	100	100	100	100	77	95.3%
Z.AI GLM 5 Turbo	100	100	100	100	76	95.2%
MoonshotAI: Kimi K2.5	100	100	100	100	74	94.8%
Qwen 3.5 9B	100	100	100	100	74	94.7%
Qwen 3.5 397B A17B	100	100	100	100	65	92.9%
Aion 2.0	100	100	100	100	65	92.9%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	63	92.6%
Gemini 3.1 Pro (Preview)	100	100	93	91	71	91.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	53	90.6%
MiniMax M2.7	100	100	94	81	78	90.5%
Gemma 3 4B	100	100	100	97	54	90.1%
ByteDance Seed 2.0 Lite	100	100	100	100	46	89.2%
Gemini 2.5 Pro	100	100	99	83	60	88.3%
Gemini 3.1 Flash Lite (Preview)	100	100	100	73	69	88.3%
Claude Sonnet 4	100	100	100	100	37	87.3%
Writer: Palmyra X5	100	100	100	100	35	87.1%
Claude Haiku 4.5	100	100	100	86	49	87.0%
Z.AI GLM 4.5	100	100	100	100	33	86.7%
Claude 3 Haiku	100	100	100	100	26	85.1%
Mistral Small 4 (Reasoning)	100	100	100	100	23	84.5%
Gemini 2.5 Flash	100	100	100	96	17	82.7%
Mistral Medium 3.1	100	100	100	98	10	81.8%
Gemini 3 Flash (Preview)	100	100	82	73	53	81.6%
GPT-4o, May 13th (temp=0)	100	100	100	79	28	81.3%
Stealth: Healer Alpha	100	100	100	65	40	81.0%
GPT-5.4 Nano (Reasoning)	100	100	100	61	41	80.4%
Qwen 3.5 122B	100	100	100	100	0	80.0%
Mistral Large	100	100	100	100	0	80.0%
Mistral Small 3.2 24B	100	100	100	100	0	80.0%
Z.AI GLM 4.6	100	100	83	59	54	79.2%
Claude Opus 4.5	100	100	100	93	0	78.6%
Qwen 3.5 27B	100	100	95	94	0	77.8%
Mistral Large 3	100	100	100	77	0	75.5%
ByteDance Seed 2.0 Mini	100	100	100	54	12	73.2%
Gemini 3 Pro (Preview)	100	100	96	50	18	72.8%
Rocinante 12B	100	100	100	63	0	72.6%
Claude 3.5 Sonnet	100	100	100	44	18	72.4%
Arcee AI: Trinity Large (Preview)	100	100	100	58	0	71.6%
Gemma 3 27B	100	100	68	55	31	70.9%
Stealth: Hunter Alpha	100	100	90	47	0	67.3%
MiniMax M2.5	100	100	100	21	11	66.2%
Ministral 3 14B	100	100	100	27	0	65.4%
DeepSeek V3.1	100	100	100	26	0	65.3%
Qwen 3.5 35B	100	71	59	53	27	62.1%
Claude Sonnet 4.6	100	100	100	0	0	60.0%
Claude 3.5 Haiku	100	100	100	0	0	60.0%
Ministral 3B	100	100	100	0	0	60.0%
Gemini 3 Flash (Preview, Reasoning)	100	83	63	52	0	59.6%
Llama 3.1 70B	100	100	77	20	0	59.3%
Ministral 8B	100	100	48	40	0	57.6%
Qwen 3.5 Flash	100	100	73	0	0	54.6%
Inception Mercury	100	79	77	0	0	51.1%
Gemma 3 12B	100	100	32	10	2	48.7%
Mistral Small Creative	100	76	54	2	0	46.4%
Z.AI GLM 4.7 Flash	100	60	34	16	0	41.9%
Z.AI GLM 4.7	42	25	5	0	0	14.5%

▼

Horror: alone in an eerie place at night

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	99	99.8%
Mistral Large	100	100	100	100	96	99.2%
GPT-5.4 Mini	100	100	100	100	92	98.4%
Stealth: Aurora Alpha	100	100	100	99	89	97.7%
Cohere Command R+ (Aug. 2024)	100	100	100	100	88	97.6%
GPT-4o, May 13th (temp=0)	100	100	100	100	88	97.5%
Claude 3.7 Sonnet	100	100	100	94	93	97.4%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	86	97.1%
Mistral Small 4	100	100	99	97	89	97.1%
WizardLM 2 8x22b	100	100	100	100	84	96.7%
GPT-4.1	100	100	100	100	83	96.6%
GPT-5.4 Mini (Reasoning)	100	100	100	94	87	96.2%
Hermes 3 70B	100	100	100	100	81	96.2%
GPT-5 Mini	100	100	100	100	80	96.0%
Claude Opus 4	100	100	100	100	79	95.9%
Inception Mercury 2	100	100	100	89	86	94.9%
GPT-5.4 Nano	100	100	100	100	72	94.4%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	66	93.3%
GPT-5.4 (Reasoning)	100	100	100	88	78	93.3%
Qwen 2.5 72B	100	100	100	84	81	92.9%
DeepSeek V3 (2025-03-24)	100	100	100	92	71	92.6%
Claude Sonnet 4.5	100	100	98	97	67	92.3%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	88	70	91.6%
GPT-4o Mini (temp=1)	100	100	100	100	51	90.1%
Grok 4.20 (Beta)	100	100	100	100	50	90.0%
Claude Opus 4.5	100	100	85	78	68	86.1%
Gemma 3 12B	100	100	91	82	50	84.5%
MoonshotAI: Kimi K2.5	100	100	100	100	19	83.7%
Mistral Large 3	100	100	100	86	29	83.1%
Nemotron 3 Nano	100	100	100	64	48	82.3%
Mistral Medium 3.1	100	100	100	95	16	82.1%
ByteDance Seed 2.0 Mini	100	100	100	67	42	81.8%
Llama 3.1 8B	100	100	100	100	5	81.1%
DeepSeek-V2 Chat	100	100	100	77	28	80.9%
Mistral Small 3.2 24B	100	100	100	100	0	80.0%
Llama 3.1 70B	100	100	100	100	0	80.0%
Arcee AI: Trinity Mini	100	100	100	100	0	80.0%
Mistral Small 4 (Reasoning)	100	96	88	77	36	79.5%
Qwen 3 32B	100	100	100	72	22	78.7%
Qwen3 235B A22B Instruct 2507	100	100	96	47	46	77.8%
Ministral 3 8B	100	100	100	74	13	77.4%
Grok 4.20 (Beta, Reasoning)	100	95	90	51	45	76.0%
Gemini 2.5 Flash	100	100	76	72	26	74.7%
GPT-5.4 Nano (Reasoning)	100	100	86	64	23	74.6%
Claude 3.5 Sonnet	100	100	94	56	21	74.3%
Writer: Palmyra X5	100	100	100	54	17	74.2%
Qwen 3.5 397B A17B	100	92	72	53	52	74.0%
Z.AI GLM 4.6	100	100	87	48	19	70.8%
DeepSeek V3.2	100	79	79	69	25	70.5%
ByteDance Seed 2.0 Lite	100	100	87	62	0	69.9%
ByteDance Seed 1.6 Flash	100	100	80	62	7	69.9%
Z.AI GLM 4.5	100	94	91	48	7	68.1%
Mistral Small Creative	100	100	100	31	0	66.1%
Stealth: Hunter Alpha	100	82	74	47	18	64.2%
Rocinante 12B	100	100	100	14	0	62.8%
Qwen 3.5 Plus (2026-02-15)	100	87	78	46	1	62.6%
Gemini 2.5 Pro	100	81	75	48	7	62.2%
Aion 2.0	100	100	94	15	0	61.7%
Gemini 2.5 Flash (Reasoning)	100	96	82	9	2	57.8%
Gemini 2.5 Flash Lite	100	100	89	0	0	57.8%
Mistral Large 2	100	100	53	28	2	56.7%
MiniMax M2.5	100	88	72	21	0	56.1%
Claude Sonnet 4	100	83	53	29	14	55.9%
Claude Sonnet 4.6	100	100	46	32	0	55.5%
DeepSeek V3.1	100	100	38	37	0	55.0%
Claude Haiku 4.5	100	98	71	0	0	53.8%
ByteDance Seed 1.6	100	90	79	0	0	53.7%
Gemini 2.5 Flash Lite (Reasoning)	100	85	72	10	0	53.4%
Gemma 3 4B	100	100	39	23	0	52.6%
Ministral 3 3B	100	74	68	5	0	49.3%
GPT-4.1 Nano	100	100	43	0	0	48.7%
Z.AI GLM 5 Turbo	100	60	48	20	0	45.6%
Ministral 8B	100	100	13	0	0	42.5%
Claude Sonnet 4.6 (Reasoning)	100	100	12	0	0	42.4%
Claude Opus 4.6 (Reasoning)	79	75	38	0	0	38.4%
Ministral 3 14B	100	31	31	27	0	37.9%
Qwen 3.5 9B	100	72	11	0	0	36.6%
Z.AI GLM 5	100	67	4	0	0	34.1%
Z.AI GLM 4.7	100	57	8	0	0	32.9%
Claude Opus 4.6	89	62	13	0	0	32.8%
MiniMax M2.7	90	24	23	15	9	32.4%
Stealth: Healer Alpha	69	50	29	12	0	32.1%
Inception Mercury	100	54	0	0	0	30.8%
Qwen 3.5 Flash	100	49	0	0	0	29.8%
Gemini 3.1 Flash Lite (Preview)	73	40	28	0	0	28.2%
Gemini 3 Flash (Preview)	59	46	26	0	0	26.3%
Gemini 3 Pro (Preview)	100	29	0	0	0	25.7%
Z.AI GLM 4.7 Flash	77	17	0	0	0	18.9%
Gemini 3.1 Pro (Preview)	69	12	6	0	0	17.4%
Claude 3.5 Haiku	41	41	4	0	0	17.3%
Qwen 3.5 122B	63	0	0	0	0	12.5%
Gemma 3 27B	50	8	3	0	0	12.3%
Ministral 3B	61	0	0	0	0	12.2%
Qwen 3.5 27B	54	0	0	0	0	10.9%
Qwen 3.5 35B	47	5	0	0	0	10.4%
Gemini 3 Flash (Preview, Reasoning)	30	0	0	0	0	6.0%

▼

Literary fiction: old friends reunite

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
Ministral 3 14B	100	100	100	100	100	100.0%
GPT-4.1 Nano	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	99.9%
GPT-5.1	100	100	100	100	97	99.3%
Qwen 3 32B	100	100	100	100	93	98.7%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	92	98.5%
Gemma 3 27B	100	100	100	100	90	97.9%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	87	97.4%
Gemma 3 12B	100	100	100	100	84	96.7%
Mistral NeMO	100	100	100	100	84	96.7%
Grok 4.20 (Beta)	100	100	100	100	79	95.8%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	72	94.5%
Claude 3.7 Sonnet	100	100	100	98	73	94.4%
Gemini 2.5 Flash Lite	100	100	100	100	67	93.3%
Qwen 3.5 Flash	100	100	98	84	83	93.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	65	93.0%
Writer: Palmyra X5	100	100	100	100	65	93.0%
Stealth: Hunter Alpha	100	100	100	100	64	92.8%
ByteDance Seed 2.0 Mini	100	100	100	97	66	92.7%
Rocinante 12B	100	100	100	100	60	92.0%
Gemma 3 4B	100	100	100	100	58	91.6%
GPT-5.4 Nano (Reasoning)	100	100	100	100	58	91.6%
Aion 2.0	100	100	100	100	58	91.5%
Claude Haiku 4.5	100	100	90	84	79	90.7%
Arcee AI: Trinity Large (Preview)	100	100	100	100	53	90.5%
Hermes 3 405B	100	100	100	100	51	90.1%
Ministral 8B	100	100	100	100	48	89.7%
Llama 3.1 70B	100	100	100	100	44	88.8%
Qwen 3.5 397B A17B	100	100	100	100	43	88.6%
Claude 3.5 Haiku	100	100	100	100	43	88.6%
Gemini 2.5 Flash	100	100	100	100	43	88.6%
Stealth: Healer Alpha	100	100	100	79	62	88.2%
GPT-5.4 Mini	100	100	100	100	38	87.6%
Claude Opus 4.5	100	100	100	100	34	86.8%
DeepSeek-V2 Chat	100	100	100	100	31	86.2%
Claude Opus 4.6	100	100	100	76	53	85.9%
Mistral Large 3	100	100	100	100	29	85.8%
Qwen 3.5 122B	100	100	94	87	36	83.5%
Gemini 3 Pro (Preview)	100	100	91	64	61	83.1%
Qwen 3.5 27B	100	100	100	98	16	82.9%
MoonshotAI: Kimi K2.5	100	100	75	69	69	82.7%
Claude Opus 4	100	100	100	67	42	81.8%
Claude 3.5 Sonnet	100	100	100	100	6	81.2%
Gemini 2.5 Pro	100	100	84	66	56	81.1%
DeepSeek V3 (2024-12-26)	100	100	100	89	16	81.0%
Mistral Large	100	100	100	64	37	80.1%
Ministral 3 3B	100	100	100	65	35	80.0%
Z.AI GLM 4.7	100	100	100	99	0	79.9%
DeepSeek V3.2	100	100	100	58	40	79.7%
ByteDance Seed 2.0 Lite	100	100	96	95	2	78.5%
Mistral Small 4	100	100	100	53	35	77.8%
Z.AI GLM 5 Turbo	100	100	96	65	27	77.5%
MiniMax M2.7	100	95	81	70	40	77.2%
Mistral Medium 3.1	100	100	100	59	18	75.5%
Z.AI GLM 5	100	100	90	82	3	74.9%
Claude Opus 4.6 (Reasoning)	100	100	91	82	0	74.7%
Ministral 3B	100	79	66	64	62	74.0%
Z.AI GLM 4.7 Flash	100	100	100	61	0	72.1%
Qwen 3.5 35B	100	100	100	56	0	71.3%
Mistral Large 2	100	100	86	33	17	67.2%
Gemini 3 Flash (Preview)	100	100	64	33	20	63.4%
MiniMax M2.5	100	96	63	53	0	62.6%
DeepSeek V3.1	100	100	41	26	18	57.1%
Claude Sonnet 4.6 (Reasoning)	100	100	72	11	0	56.7%
Gemini 3 Flash (Preview, Reasoning)	100	85	46	35	8	54.9%
Ministral 3 8B	100	100	33	0	0	46.5%
Qwen 3.5 9B	100	53	23	15	12	40.4%
Claude Sonnet 4.6	85	46	0	0	0	26.0%
Z.AI GLM 4.6	59	38	25	3	0	25.0%

▼

Mystery: examining a crime scene

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	98	99.7%
Arcee AI: Trinity Mini	100	100	100	100	96	99.2%
GPT-4.1 Mini	100	100	100	100	95	99.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	89	97.8%
DeepSeek V3 (2024-12-26)	100	100	100	100	84	96.8%
Llama 3.1 70B	100	100	100	100	80	96.0%
Claude Sonnet 4.5	100	100	100	100	79	95.8%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	75	94.9%
Cohere Command R+ (Aug. 2024)	100	100	100	100	74	94.8%
Qwen 2.5 72B	100	100	100	100	69	93.9%
GPT-5.4 Mini (Reasoning)	100	100	100	96	73	93.9%
Mistral NeMO	100	100	100	88	80	93.6%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	68	93.6%
Mistral Small 4 (Reasoning)	100	100	100	92	74	93.3%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	63	92.6%
Gemma 3 4B	100	100	100	100	60	92.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	53	90.6%
GPT-4o Mini (temp=1)	100	100	100	100	52	90.4%
ByteDance Seed 1.6	100	100	100	96	46	88.4%
ByteDance Seed 2.0 Lite	100	100	100	80	62	88.3%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	41	88.1%
GPT-4o Mini (temp=0)	100	100	100	92	37	85.8%
Stealth: Aurora Alpha	100	100	100	80	43	84.6%
GPT-5.4 Mini	100	100	100	84	37	84.4%
Gemma 3 27B	100	100	100	97	22	83.9%
ByteDance Seed 2.0 Mini	100	100	100	83	36	83.8%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	0	80.0%
MiniMax M2.7	100	100	100	100	0	80.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	0	80.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	0	80.0%
Hermes 3 70B	100	100	100	100	0	80.0%
Gemini 3.1 Pro (Preview)	100	92	76	65	59	78.4%
Qwen 3.5 Plus (2026-02-15)	100	100	99	70	13	76.3%
Mistral Medium 3.1	100	100	100	62	17	75.8%
Claude Opus 4.5	100	98	70	59	50	75.5%
Mistral Small Creative	100	100	100	59	14	74.6%
Z.AI GLM 4.5	100	100	96	48	21	73.1%
Claude 3.5 Sonnet	100	100	83	79	0	72.4%
Claude 3 Haiku	100	100	100	40	18	71.6%
Stealth: Healer Alpha	100	100	100	57	0	71.4%
ByteDance Seed 1.6 Flash	100	100	100	56	0	71.3%
Grok 4.20 (Beta)	100	83	77	46	46	70.6%
Gemma 3 12B	100	100	100	52	0	70.4%
Mistral Large 3	100	100	73	40	36	69.8%
Qwen 3.5 Flash	100	100	100	37	0	67.5%
WizardLM 2 8x22b	100	100	73	64	0	67.5%
DeepSeek V3.2	100	90	88	54	0	66.4%
GPT-4.1 Nano	100	100	77	55	0	66.3%
Mistral Small 3.2 24B	100	89	78	31	30	65.6%
Aion 2.0	100	100	100	25	0	65.1%
Claude Opus 4.6	100	100	61	43	17	64.0%
Mistral Small 4	100	79	72	63	0	62.6%
Claude Opus 4	100	100	100	6	4	62.0%
Claude Opus 4.6 (Reasoning)	100	100	86	23	0	61.7%
Mistral Large	100	100	53	47	0	60.0%
Qwen 3.5 9B	100	100	100	0	0	60.0%
Inception Mercury	100	100	100	0	0	60.0%
Ministral 3 3B	100	100	82	8	0	58.0%
Claude Sonnet 4	100	80	68	41	0	57.8%
Z.AI GLM 5	100	100	53	7	0	52.2%
Mistral Large 2	100	81	52	23	0	51.2%
Ministral 8B	100	86	53	17	0	51.1%
Gemini 3 Pro (Preview)	100	79	45	31	0	50.9%
Ministral 3 14B	100	77	75	0	0	50.5%
Z.AI GLM 4.7	100	68	63	5	0	47.1%
DeepSeek V3.1	100	99	26	11	0	47.1%
Gemini 3 Flash (Preview, Reasoning)	100	85	32	11	0	45.8%
Claude Haiku 4.5	100	100	26	0	0	45.1%
MiniMax M2.5	100	72	52	0	0	44.9%
Stealth: Hunter Alpha	100	96	24	0	0	44.0%
Claude Sonnet 4.6	100	49	42	28	0	43.9%
Gemini 3.1 Flash Lite (Preview)	78	45	45	41	0	41.6%
Ministral 3B	100	79	29	0	0	41.6%
Ministral 3 8B	100	100	0	0	0	40.0%
Qwen 3.5 122B	100	50	49	0	0	39.9%
Qwen 3.5 397B A17B	77	75	14	14	0	35.9%
Z.AI GLM 4.7 Flash	100	68	0	0	0	33.6%
Z.AI GLM 5 Turbo	62	39	34	0	0	27.0%
Gemini 2.5 Pro	96	19	12	0	0	25.4%
Qwen 3.5 27B	87	39	0	0	0	25.2%
Qwen 3.5 35B	56	51	0	0	0	21.4%
Claude Sonnet 4.6 (Reasoning)	100	0	0	0	0	20.0%
Z.AI GLM 4.6	58	0	0	0	0	11.6%
Gemini 3 Flash (Preview)	1	0	0	0	0	0.2%

▼

Romance: separated couple reunites

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	98	99.6%
Qwen 3.5 122B	100	100	100	100	97	99.4%
Grok 4.20 (Beta)	100	100	100	100	93	98.6%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	92	98.4%
GPT-5.4 Nano (Reasoning)	100	100	100	100	91	98.1%
Stealth: Healer Alpha	100	100	100	100	88	97.7%
Stealth: Hunter Alpha	100	100	100	99	89	97.7%
o4 Mini	100	100	100	100	86	97.3%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	83	96.6%
GPT-5.4 Nano	100	100	100	91	91	96.6%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	98	82	96.0%
Hermes 3 405B	100	100	100	100	71	94.1%
Gemma 3 27B	100	100	100	100	70	94.0%
GPT-4.1 Nano	100	100	100	100	67	93.3%
Ministral 3 14B	100	100	98	90	77	93.2%
Z.AI GLM 4.5	100	100	100	100	63	92.6%
Claude 3.7 Sonnet	100	100	100	100	63	92.5%
Mistral Large	100	100	100	94	66	91.9%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	57	91.3%
Qwen 3.5 Flash	100	100	100	100	56	91.1%
Claude 3.5 Sonnet	100	100	100	87	65	90.5%
Qwen 2.5 72B	100	100	100	100	52	90.4%
Claude Sonnet 4.5	100	100	100	100	47	89.3%
Mistral Large 2	100	100	100	100	44	88.8%
Gemini 3.1 Pro (Preview)	100	100	100	100	43	88.6%
Arcee AI: Trinity Large (Preview)	100	100	100	96	46	88.5%
Z.AI GLM 5 Turbo	100	100	100	100	41	88.3%
Qwen 3.5 35B	100	100	100	79	60	87.9%
Nemotron 3 Nano	100	100	100	100	37	87.5%
Mistral Medium 3.1	100	100	100	87	50	87.3%
Gemini 2.5 Pro	100	100	100	70	65	87.0%
Qwen 3.5 397B A17B	100	100	100	100	27	85.5%
Inception Mercury	100	100	100	100	26	85.1%
Claude 3.5 Haiku	100	100	100	100	25	84.9%
Rocinante 12B	100	100	100	100	25	84.9%
MoonshotAI: Kimi K2.5	100	100	100	79	45	84.7%
Gemini 3 Flash (Preview, Reasoning)	100	94	82	77	63	83.2%
Llama 3.1 Nemotron 70B	100	100	100	100	15	83.0%
Claude Opus 4.5	100	100	93	63	58	82.9%
Aion 2.0	100	100	97	83	27	81.6%
Claude Opus 4	100	100	100	73	34	81.4%
Hermes 3 70B	100	100	100	100	6	81.2%
Z.AI GLM 5	100	100	100	95	9	80.8%
Gemini 3 Flash (Preview)	100	100	100	88	16	80.7%
DeepSeek-V2 Chat	100	100	97	55	51	80.6%
Claude Opus 4.6	100	100	99	99	4	80.4%
DeepSeek V3.1	100	100	100	100	0	80.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	0	80.0%
Mistral NeMO	100	100	100	100	0	80.0%
Ministral 3 8B	100	100	100	94	0	78.8%
DeepSeek V3.2	100	100	100	91	0	78.3%
Z.AI GLM 4.7	100	100	100	88	0	77.7%
Gemini 2.5 Flash (Reasoning)	100	100	95	77	12	76.8%
Mistral Large 3	100	100	100	84	0	76.7%
MiniMax M2.7	100	100	100	63	18	76.1%
Z.AI GLM 4.6	100	100	100	53	25	75.6%
Ministral 8B	100	100	100	71	0	74.2%
DeepSeek V3 (2024-12-26)	100	83	71	68	37	71.8%
Z.AI GLM 4.7 Flash	100	100	81	63	2	69.4%
Gemini 3.1 Flash Lite (Preview)	100	100	64	36	33	66.7%
Mistral Small Creative	100	100	100	19	0	63.7%
Claude Opus 4.6 (Reasoning)	100	78	67	46	17	61.5%
Gemma 3 4B	100	100	73	25	0	59.5%
Claude Haiku 4.5	100	100	46	41	0	57.4%
Qwen 3.5 27B	100	100	36	28	0	52.8%
Claude Sonnet 4.6 (Reasoning)	100	100	44	13	0	51.5%
Gemini 3 Pro (Preview)	100	53	37	23	2	43.0%
Ministral 3B	100	85	0	0	0	37.0%
Claude Sonnet 4.6	82	80	2	0	0	32.9%
MiniMax M2.5	86	45	26	0	0	31.5%

▼

Thriller: chase through city streets

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	98	99.6%
Arcee AI: Trinity Large (Preview)	100	100	100	100	96	99.3%
Grok 4 Fast	100	100	100	100	96	99.2%
Claude 3.5 Sonnet	100	100	100	100	93	98.7%
GPT-4o Mini (temp=0)	100	100	100	100	88	97.7%
GPT-5.4 Nano	100	100	100	100	87	97.5%
ByteDance Seed 1.6 Flash	100	100	100	100	80	96.0%
GPT-5.4 (Reasoning)	100	100	100	100	79	95.8%
Mistral Large 2	100	100	100	94	80	94.8%
Claude Sonnet 4.5	100	100	100	94	74	93.7%
Qwen3 235B A22B Instruct 2507	100	100	100	100	67	93.5%
Claude Sonnet 4	100	100	100	100	67	93.3%
Arcee AI: Trinity Mini	100	100	100	100	65	93.1%
Gemini 3.1 Pro (Preview)	100	100	100	84	81	93.0%
Llama 3.1 Nemotron 70B	100	100	100	100	57	91.3%
GPT-5.4 Mini (Reasoning)	100	100	100	84	72	91.2%
GPT-4o, May 13th (temp=0)	100	100	100	100	54	90.8%
Claude 3 Haiku	100	100	100	81	72	90.5%
GPT-5.4 Mini (Reasoning, Low)	100	100	95	86	69	89.9%
Rocinante 12B	100	100	100	100	48	89.7%
Mistral Medium 3.1	100	100	100	100	47	89.5%
Mistral Small 4 (Reasoning)	100	100	100	74	73	89.5%
DeepSeek V3.2	100	100	96	89	59	88.9%
Claude 3.7 Sonnet	100	100	100	100	43	88.5%
Hermes 3 405B	100	100	100	100	40	88.1%
Cohere Command R+ (Aug. 2024)	100	100	100	100	33	86.7%
Qwen 2.5 72B	100	100	100	86	45	86.2%
Nemotron 3 Super	100	100	100	75	53	85.6%
GPT-5.4 (Reasoning, Low)	100	100	100	72	42	82.8%
MoonshotAI: Kimi K2.5	100	100	88	78	35	80.2%
Mistral Small 4	100	100	100	100	0	80.0%
GPT-4.1 Nano	100	100	100	100	0	80.0%
Inception Mercury	100	100	100	95	0	79.1%
Claude Opus 4	100	100	100	94	0	78.8%
Grok 4.20 (Beta, Reasoning)	100	100	76	69	48	78.5%
GPT-4o, May 13th (temp=1)	100	100	100	91	0	78.1%
Gemma 3 4B	100	100	88	66	30	76.8%
MiniMax M2.7	100	100	100	44	37	76.4%
Mistral Small 3.2 24B	100	100	100	44	31	74.8%
Z.AI GLM 4.5	100	100	84	78	0	72.5%
Qwen 3.5 Plus (2026-02-15)	100	100	100	61	0	72.2%
Grok 4.20 (Beta)	100	89	74	60	21	68.9%
Qwen 3.5 9B	100	100	74	69	0	68.7%
Llama 3.1 70B	100	100	85	37	16	67.7%
Qwen 3.5 397B A17B	100	97	76	44	20	67.4%
Gemini 2.5 Flash Lite	100	100	53	53	21	65.5%
Mistral NeMO	100	100	72	42	0	63.0%
GPT-5.4 Mini	100	100	90	21	0	62.1%
Gemini 3 Flash (Preview)	100	100	66	41	0	61.2%
Qwen 3.5 Flash	100	100	100	2	0	60.3%
Claude Opus 4.6 (Reasoning)	100	90	83	27	0	59.9%
Mistral Small Creative	100	100	94	0	0	58.8%
Mistral Large	100	72	49	46	23	57.9%
ByteDance Seed 2.0 Mini	100	94	53	41	0	57.8%
MiniMax M2.5	100	80	57	44	0	56.2%
Ministral 8B	100	71	56	32	19	55.6%
Mistral Large 3	100	75	62	39	1	55.3%
Stealth: Healer Alpha	100	81	56	22	14	54.6%
Claude Opus 4.5	100	100	55	2	0	51.5%
Gemini 2.5 Flash	100	100	49	0	0	49.9%
Gemma 3 12B	100	70	63	14	0	49.3%
Claude Sonnet 4.6	100	100	40	0	0	48.0%
Gemini 3.1 Flash Lite (Preview)	100	95	39	5	0	47.9%
Claude 3.5 Haiku	100	100	33	0	0	46.7%
DeepSeek V3.1	100	84	29	15	0	45.5%
Qwen 3.5 35B	82	77	52	14	0	45.0%
Stealth: Hunter Alpha	100	62	41	17	0	44.0%
Z.AI GLM 5 Turbo	100	76	31	13	0	44.0%
Gemini 2.5 Pro	100	98	8	0	0	41.3%
Z.AI GLM 5	100	92	10	2	0	40.8%
Qwen 3 32B	100	86	11	0	0	39.4%
Gemini 3 Pro (Preview)	100	50	38	9	0	39.4%
Claude Opus 4.6	91	53	51	0	0	39.1%
Ministral 3B	100	66	25	0	0	38.1%
Qwen 3.5 27B	100	59	26	0	0	37.1%
Aion 2.0	91	72	21	0	0	36.9%
Gemini 3 Flash (Preview, Reasoning)	100	48	26	0	0	34.9%
Claude Sonnet 4.6 (Reasoning)	79	48	41	0	0	33.6%
Ministral 3 14B	59	36	31	22	3	30.4%
Claude Haiku 4.5	100	23	21	0	0	28.7%
Z.AI GLM 4.6	100	33	0	0	0	26.5%
Ministral 3 8B	64	60	7	0	0	26.3%
Z.AI GLM 4.7 Flash	67	56	0	0	0	24.7%
Ministral 3 3B	47	43	29	0	0	23.8%
Qwen 3.5 122B	88	28	0	0	0	23.3%
Z.AI GLM 4.7	60	27	8	0	0	18.9%
Gemma 3 27B	61	18	5	0	0	16.8%

Novelcrafter Default Prompt

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Aion 2.0	100	100	100	100	100	100.0%
MiniMax M2.7	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
GPT-4.1 Nano	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	99.9%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	99	99.9%
GPT-4o, May 13th (temp=1)	100	100	100	100	99	99.8%
Claude Opus 4	100	100	100	100	97	99.5%
Claude 3.7 Sonnet	100	100	100	100	95	99.1%
Mistral Large 3	100	100	100	100	95	98.9%
GPT-5 Nano	100	100	100	100	94	98.8%
WizardLM 2 8x22b	100	100	100	100	92	98.5%
GPT-4o, May 13th (temp=0)	100	100	100	100	91	98.1%
Mistral Small Creative	100	100	100	100	87	97.3%
Mistral Large 2	100	100	100	100	86	97.3%
Gemini 3 Flash (Preview)	100	100	100	100	83	96.7%
Claude Sonnet 4	100	100	100	100	83	96.6%
Gemma 3 4B	100	100	100	100	80	96.0%
Stealth: Hunter Alpha	100	100	100	100	66	93.2%
DeepSeek V3.1	100	100	100	100	65	93.0%
Gemini 3 Pro (Preview)	100	100	100	100	64	92.9%
Arcee AI: Trinity Large (Preview)	100	100	100	100	63	92.6%
Gemini 2.5 Pro	100	100	100	100	55	91.0%
MiniMax M2.5	100	100	100	100	54	90.9%
Cohere Command R+ (Aug. 2024)	100	100	100	100	52	90.4%
Claude Opus 4.6	100	100	96	87	61	89.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	41	88.3%
Claude Opus 4.5	100	100	100	94	44	87.6%
Rocinante 12B	100	100	100	95	40	87.1%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	34	86.9%
Z.AI GLM 4.6	100	100	100	100	28	85.6%
Stealth: Healer Alpha	100	100	100	100	25	85.1%
Gemma 3 12B	100	100	100	100	25	84.9%
Mistral Medium 3.1	100	100	100	71	41	82.4%
Claude Sonnet 4.6	100	100	100	100	11	82.2%
Qwen 3.5 35B	100	100	100	100	0	80.0%
Mistral Small 3.2 24B	100	100	100	100	0	80.0%
Llama 3.1 Nemotron 70B	100	100	100	100	0	80.0%
Claude 3.5 Haiku	100	100	100	96	0	79.2%
Ministral 3 3B	100	100	100	89	0	77.8%
Ministral 3 14B	100	94	77	71	40	76.4%
Claude Haiku 4.5	100	100	78	68	0	69.1%
Z.AI GLM 4.7 Flash	100	100	52	0	0	50.4%

▼

Horror: alone in an eerie place at night

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	95	99.1%
Claude Sonnet 4.6	100	100	100	100	86	97.1%
Mistral NeMO	100	100	100	100	83	96.5%
Nemotron 3 Nano	100	100	100	100	82	96.5%
Gemini 3 Pro (Preview)	100	100	100	100	80	96.0%
Claude Sonnet 4	100	100	100	100	80	96.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	79	95.8%
ByteDance Seed 2.0 Lite	100	100	100	100	74	94.8%
GPT-4o, May 13th (temp=0)	100	100	100	100	74	94.7%
Gemini 2.5 Flash Lite	100	100	97	88	88	94.5%
Mistral Small 4 (Reasoning)	100	100	97	93	81	94.3%
Mistral Small 4	100	100	100	98	72	94.0%
Claude 3.5 Sonnet	100	100	100	87	82	93.8%
MiniMax M2.7	100	100	100	100	67	93.5%
Aion 2.0	100	100	100	87	77	92.9%
Arcee AI: Trinity Large (Preview)	100	100	100	100	59	91.8%
Hermes 3 405B	100	100	100	100	58	91.6%
Llama 3.1 70B	100	100	100	95	62	91.4%
Ministral 8B	100	100	100	100	57	91.3%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	56	91.2%
ByteDance Seed 2.0 Mini	100	100	100	83	72	90.9%
Ministral 3 8B	100	100	100	100	53	90.7%
Gemma 3 12B	100	100	100	79	74	90.6%
Arcee AI: Trinity Mini	100	100	100	100	51	90.1%
Claude Opus 4.6	100	100	99	91	60	90.0%
Gemini 2.5 Pro	100	100	100	100	50	90.0%
Rocinante 12B	100	100	100	89	58	89.4%
Cohere Command R+ (Aug. 2024)	100	100	100	100	44	88.8%
Llama 3.1 Nemotron 70B	100	100	100	81	63	88.7%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	84	59	88.6%
Claude Opus 4.6 (Reasoning)	100	100	100	97	44	88.2%
Qwen 3 32B	100	100	100	100	38	87.7%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	36	87.3%
GPT-4.1 Nano	100	100	100	85	50	87.0%
Mistral Large	100	100	100	100	28	85.5%
Qwen 3.5 35B	100	100	100	72	53	85.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	88	34	84.4%
DeepSeek-V2 Chat	100	100	73	69	69	82.2%
Ministral 3 3B	100	100	100	100	7	81.4%
Claude Haiku 4.5	100	100	84	68	52	80.8%
Inception Mercury	100	100	100	100	0	80.0%
Mistral Small 3.2 24B	100	100	100	100	0	80.0%
Ministral 3B	100	100	100	99	0	79.8%
ByteDance Seed 1.6	100	100	100	73	21	78.9%
Claude Opus 4	100	100	100	79	12	78.1%
DeepSeek V3 (2024-12-26)	100	100	100	84	0	76.7%
Claude Opus 4.5	100	100	100	50	32	76.5%
Qwen 2.5 72B	100	100	100	80	0	76.0%
Ministral 3 14B	100	100	100	63	0	72.5%
Gemma 3 27B	100	100	100	50	0	69.9%
Claude 3 Haiku	100	87	73	60	27	69.3%
Z.AI GLM 5	100	100	80	32	32	68.6%
Gemma 3 4B	100	100	100	34	0	66.8%
Claude Sonnet 4.6 (Reasoning)	100	100	100	29	0	65.8%
Stealth: Hunter Alpha	100	100	77	49	0	65.3%
Stealth: Healer Alpha	100	76	71	69	0	63.1%
Gemini 3 Flash (Preview)	100	100	73	42	0	63.0%
Z.AI GLM 4.7 Flash	100	80	72	52	0	60.8%
MiniMax M2.5	100	100	75	0	0	55.0%
DeepSeek V3.1	100	77	66	7	0	50.0%
WizardLM 2 8x22b	100	86	57	6	0	49.7%
Claude 3.5 Haiku	100	100	29	0	0	45.8%
DeepSeek V3.2	100	48	32	24	0	40.8%
Z.AI GLM 4.7	100	67	0	0	0	33.3%
Z.AI GLM 4.6	74	59	28	0	0	32.3%

▼

Literary fiction: old friends reunite

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Qwen 3.5 35B	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Stealth: Hunter Alpha	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
GPT-4.1 Nano	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	98	99.7%
GPT-5 Mini	100	100	100	100	96	99.2%
Claude Opus 4.6 (Reasoning)	100	100	100	100	91	98.2%
ByteDance Seed 1.6	100	100	100	100	88	97.7%
DeepSeek-V2 Chat	100	100	100	100	88	97.5%
Z.AI GLM 4.7 Flash	100	100	100	100	87	97.4%
DeepSeek V3 (2025-03-24)	100	100	100	100	85	97.0%
Claude Opus 4.6	100	100	100	100	82	96.4%
MiniMax M2.7	100	100	100	100	80	96.1%
ByteDance Seed 2.0 Lite	100	100	100	100	70	94.0%
Stealth: Healer Alpha	100	100	100	89	79	93.7%
Gemma 3 4B	100	100	100	100	68	93.5%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	65	93.1%
Aion 2.0	100	100	100	84	80	92.6%
Mistral Small 4 (Reasoning)	100	100	100	87	76	92.5%
Mistral Small 3.2 24B	100	100	100	100	61	92.2%
Claude Haiku 4.5	100	100	100	82	78	92.1%
Z.AI GLM 5	100	100	100	100	60	91.9%
DeepSeek V3.1	100	100	100	100	44	88.8%
Mistral Small Creative	100	100	100	98	36	86.8%
Mistral Large 2	100	100	100	87	44	86.2%
Claude Sonnet 4.5	100	100	100	100	25	84.9%
MiniMax M2.5	100	100	100	99	20	83.9%
DeepSeek V3 (2024-12-26)	100	100	100	100	7	81.4%
Mistral Large 3	100	100	100	100	0	80.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	0	80.0%
Ministral 3 3B	100	100	100	100	0	80.0%
Claude Sonnet 4.6 (Reasoning)	100	100	100	98	0	79.6%
Gemini 2.5 Pro	100	100	100	64	19	76.6%
Gemma 3 27B	100	100	100	61	21	76.4%
Mistral Large	100	100	100	77	0	75.5%
Mistral Medium 3.1	100	100	100	38	35	74.8%
ByteDance Seed 2.0 Mini	100	100	100	63	0	72.7%
Claude 3.5 Haiku	100	100	100	37	15	70.4%
Ministral 3 8B	100	100	100	47	0	69.4%
Gemini 3 Flash (Preview)	100	100	100	45	0	69.0%
Ministral 3 14B	100	100	100	30	11	68.2%
Ministral 3B	100	100	100	38	0	67.7%
Ministral 8B	100	100	100	0	0	60.0%
Z.AI GLM 4.6	100	81	76	0	0	51.3%
Claude Sonnet 4.6	100	54	0	0	0	30.8%

▼

Mystery: examining a crime scene

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
MiniMax M2.7	100	100	100	100	98	99.6%
Gemini 2.5 Flash Lite	100	100	100	100	97	99.5%
MiniMax M2.5	100	100	100	100	93	98.7%
Claude Opus 4	100	100	100	100	92	98.3%
DeepSeek V3 (2024-12-26)	100	100	100	100	91	98.1%
Z.AI GLM 5 Turbo	100	100	100	100	82	96.4%
Qwen 2.5 72B	100	100	100	100	82	96.4%
Cohere Command R+ (Aug. 2024)	100	100	100	100	82	96.4%
Arcee AI: Trinity Mini	100	100	100	100	79	95.8%
GPT-4o, Aug. 6th (temp=0)	100	100	100	92	83	94.9%
Claude Haiku 4.5	100	100	100	100	71	94.1%
Claude Opus 4.5	100	100	100	100	69	93.7%
Gemma 3 4B	100	100	100	100	67	93.3%
Gemini 3.1 Flash Lite (Preview)	100	100	100	85	81	93.1%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	62	92.4%
Gemini 2.5 Pro	100	100	100	100	60	92.0%
Mistral Small 3.2 24B	100	100	100	100	58	91.6%
Llama 3.1 70B	100	100	100	100	58	91.6%
Stealth: Healer Alpha	100	100	89	87	77	90.6%
Claude Opus 4.6	100	100	100	98	52	90.0%
Aion 2.0	100	100	100	100	48	89.7%
Ministral 3 14B	100	100	100	79	69	89.6%
Claude 3.5 Haiku	100	100	100	100	41	88.3%
Ministral 8B	100	100	100	100	39	87.9%
Stealth: Hunter Alpha	100	100	100	76	62	87.6%
Qwen 3.5 Flash	100	100	100	100	33	86.7%
WizardLM 2 8x22b	100	100	100	65	64	85.9%
Hermes 3 70B	100	100	100	100	29	85.8%
Gemini 3 Pro (Preview)	100	100	100	100	25	85.1%
GPT-4.1 Nano	100	100	100	100	23	84.7%
ByteDance Seed 1.6	100	100	100	100	20	84.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	12	82.4%
Claude 3 Haiku	100	100	100	85	22	81.5%
Writer: Palmyra X5	100	100	100	92	15	81.3%
Qwen 3.5 35B	100	100	100	100	0	80.0%
Inception Mercury	100	100	100	100	0	80.0%
Gemma 3 12B	100	100	100	100	0	80.0%
Mistral Small Creative	100	100	100	100	0	80.0%
Ministral 3 8B	100	100	100	100	0	80.0%
Rocinante 12B	100	100	100	100	0	80.0%
Gemini 3 Flash (Preview)	100	100	100	54	39	78.7%
Hermes 3 405B	100	100	100	64	25	77.7%
Qwen 3 32B	100	100	92	89	0	76.3%
Ministral 3B	100	100	87	74	17	75.7%
Mistral Large 3	100	100	84	78	0	72.3%
Z.AI GLM 4.7	100	100	84	76	0	72.0%
Arcee AI: Trinity Large (Preview)	100	100	100	35	25	72.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	57	0	71.4%
Mistral NeMO	100	93	89	57	0	67.9%
DeepSeek V3.1	100	84	81	55	0	64.0%
DeepSeek V3.2	100	91	29	14	0	46.7%
Z.AI GLM 4.6	100	83	21	0	0	40.9%
Z.AI GLM 4.7 Flash	0	0	0	0	0	0.0%

▼

Romance: separated couple reunites

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Qwen 3.5 35B	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Stealth: Healer Alpha	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	97	99.5%
Llama 3.1 70B	100	100	100	100	93	98.7%
Z.AI GLM 4.5	100	100	100	100	89	97.8%
MoonshotAI: Kimi K2.5	100	100	100	100	86	97.1%
Z.AI GLM 4.7	100	100	100	100	86	97.1%
Claude Opus 4	100	100	100	100	84	96.9%
Mistral Medium 3.1	100	100	100	100	83	96.6%
Arcee AI: Trinity Mini	100	100	100	100	83	96.6%
Stealth: Hunter Alpha	100	100	100	96	83	95.8%
Aion 2.0	100	100	100	100	77	95.3%
Ministral 3 3B	100	100	100	100	76	95.3%
ByteDance Seed 1.6 Flash	100	100	100	100	75	95.1%
Claude Sonnet 4	100	100	100	100	73	94.6%
DeepSeek V3 (2024-12-26)	100	100	100	88	82	94.1%
Claude Opus 4.6 (Reasoning)	100	100	100	100	69	93.8%
GPT-4.1 Nano	100	100	100	100	63	92.5%
Ministral 8B	100	100	100	100	57	91.4%
Z.AI GLM 4.7 Flash	100	100	100	100	46	89.2%
Z.AI GLM 5 Turbo	100	100	100	100	45	89.0%
Claude Opus 4.6	100	100	100	100	44	88.8%
ByteDance Seed 2.0 Lite	100	100	100	75	67	88.5%
Gemini 3 Pro (Preview)	100	100	100	100	37	87.5%
Mistral Small Creative	100	100	100	89	45	86.8%
MiniMax M2.7	100	100	100	100	15	83.0%
Ministral 3B	100	100	100	88	25	82.6%
Claude Opus 4.5	100	100	100	99	5	80.8%
Claude Haiku 4.5	100	100	100	100	0	80.0%
Claude 3.5 Haiku	100	100	100	100	0	80.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	0	80.0%
Ministral 3 14B	100	100	100	100	0	80.0%
Mistral NeMO	100	100	100	100	0	80.0%
DeepSeek V3.1	100	100	98	57	31	77.1%
Llama 3.1 8B	100	100	100	58	20	75.6%
Gemini 3 Flash (Preview)	100	100	100	77	0	75.4%
MiniMax M2.5	100	100	72	57	15	68.8%
Claude Sonnet 4.6 (Reasoning)	99	89	59	39	10	59.1%
Claude Sonnet 4.6	100	94	48	0	0	48.5%

▼

Thriller: chase through city streets

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
MiniMax M2.5	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Qwen 3.5 35B	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100.0%
MiniMax M2.7	100	100	100	100	98	99.6%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	93	98.7%
GPT-4o, May 13th (temp=0)	100	100	100	100	92	98.5%
Ministral 3B	100	100	100	100	91	98.1%
Gemini 2.5 Flash Lite	100	100	100	100	88	97.7%
Z.AI GLM 5	100	100	100	100	88	97.5%
Writer: Palmyra X5	100	100	100	100	84	96.7%
Claude Sonnet 4.5	100	100	100	100	82	96.3%
Claude Sonnet 4	100	100	100	100	75	95.0%
Claude 3.5 Sonnet	100	100	100	100	74	94.8%
Aion 2.0	100	100	97	94	81	94.5%
ByteDance Seed 2.0 Mini	100	100	100	100	71	94.1%
Llama 3.1 Nemotron 70B	100	100	100	100	68	93.6%
Llama 3.1 8B	100	100	100	85	81	93.2%
Gemma 3 12B	100	100	100	100	64	92.8%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	63	92.6%
Claude Opus 4	100	100	96	86	80	92.5%
Claude Opus 4.6	100	100	100	100	61	92.1%
Hermes 3 405B	100	100	100	100	56	91.2%
Arcee AI: Trinity Mini	100	100	100	83	68	90.2%
Rocinante 12B	100	100	100	75	74	89.9%
Z.AI GLM 4.7	100	98	84	81	80	88.6%
Claude Haiku 4.5	100	100	100	100	42	88.4%
Stealth: Hunter Alpha	100	100	100	100	41	88.3%
GPT-4o, May 13th (temp=1)	100	100	100	76	64	87.9%
Gemini 3 Pro (Preview)	100	100	100	89	46	86.8%
Qwen 2.5 72B	100	100	100	89	43	86.5%
DeepSeek V3.1	100	100	100	68	60	85.7%
Z.AI GLM 5 Turbo	100	100	100	69	54	84.7%
GPT-4.1 Nano	100	100	100	100	22	84.4%
Mistral Small 4 (Reasoning)	100	100	97	67	56	83.9%
Claude 3 Haiku	100	100	81	75	63	83.8%
Gemma 3 4B	100	100	100	59	53	82.5%
DeepSeek V3.2	100	100	100	79	32	82.2%
DeepSeek V3 (2025-03-24)	100	100	100	55	53	81.7%
Claude 3.5 Haiku	100	100	100	100	4	80.8%
Mistral Large	100	100	100	87	17	80.7%
Z.AI GLM 4.6	100	100	100	100	0	80.0%
Claude 3.7 Sonnet	100	100	100	100	0	80.0%
Mistral Large 3	100	100	100	94	0	78.8%
Claude Opus 4.5	100	100	86	67	24	75.4%
Gemini 2.5 Pro	100	84	82	61	48	75.1%
Cohere Command R+ (Aug. 2024)	100	100	100	73	0	74.5%
Mistral NeMO	100	88	85	58	31	72.4%
Hermes 3 70B	100	100	100	61	0	72.2%
Mistral Medium 3.1	100	100	100	58	0	71.6%
Mistral Large 2	100	100	81	74	0	71.0%
Z.AI GLM 4.7 Flash	100	97	88	69	0	70.8%
Z.AI GLM 4.5	100	100	100	50	0	69.9%
Ministral 8B	100	100	89	57	0	69.2%
Qwen 3 32B	100	100	100	36	0	67.2%
Mistral Small Creative	100	100	74	40	20	66.9%
Mistral Small 3.2 24B	100	100	100	30	0	66.0%
Llama 3.1 70B	100	100	97	29	0	65.3%
DeepSeek-V2 Chat	100	100	62	41	22	65.1%
Claude Sonnet 4.6	100	77	66	52	30	64.9%
Inception Mercury	100	100	100	0	0	60.0%
Ministral 3 8B	100	100	100	0	0	60.0%
Ministral 3 14B	100	100	92	0	0	58.3%
Claude Sonnet 4.6 (Reasoning)	100	100	47	41	0	57.6%
Claude Opus 4.6 (Reasoning)	100	100	81	0	0	56.3%
WizardLM 2 8x22b	100	100	53	28	0	56.1%
Gemini 3 Flash (Preview, Reasoning)	100	100	51	14	0	52.9%
Gemini 3 Flash (Preview)	100	63	55	43	0	52.0%
Stealth: Healer Alpha	100	100	53	0	0	50.5%

Past progressive (was/were + -ing) overuse

Overall Performance

Performance Score Distribution (Top 20)

Price-Performance Score Distribution (Top 20)

Most Stable Models (Top 20)

Top Overall Models (Top 20)

Individual Scenarios

Detailed Writing Rules

Fantasy: entering an ancient ruin

Horror: alone in an eerie place at night

Literary fiction: old friends reunite

Mystery: examining a crime scene

Romance: separated couple reunites

Thriller: chase through city streets

genre

Fantasy: entering an ancient ruin

Horror: alone in an eerie place at night

Literary fiction: old friends reunite

Mystery: examining a crime scene

Romance: separated couple reunites

Thriller: chase through city streets

Novelcrafter Default Prompt

Fantasy: entering an ancient ruin

Horror: alone in an eerie place at night

Literary fiction: old friends reunite

Mystery: examining a crime scene

Romance: separated couple reunites

Thriller: chase through city streets