Passive voice overuse

Test: Bad Writing Habits

Avg. Score

96.7%

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	Grok 4.1 Fast	99.9%	$0.0018	37.8s	99%
2	Grok 4 Fast	99.6%	$0.0017	24.1s	96%
3	o4 Mini	99.5%	$0.015	25.7s	97%
4	o4 Mini High	99.8%	$0.025	47.2s	99%
5	GPT-4o Mini (temp=1)	99.0%	$0.0012	34.8s	96%
6	Writer: Palmyra X5	98.8%	$0.011	22.0s	96%
7	Qwen3 235B A22B Instruct 2507	99.0%	$0.0011	59.2s	96%
8	Mistral Small 4 (Reasoning)	98.7%	$0.0022	30.2s	95%
9	DeepSeek V3 (2025-03-24)	98.8%	$0.0014	39.4s	94%
10	GPT-4.1	99.3%	$0.018	44.7s	95%
11	Mistral Small 4	98.2%	$0.0014	18.2s	93%
12	LFM2 24B	98.2%	$0.0002	28.4s	93%
13	Gemini 2.5 Flash (Reasoning)	98.1%	$0.011	21.5s	93%
14	Grok 4.3	98.6%	$0.0069	30.5s	92%
15	GPT-4.1 Mini	98.1%	$0.0027	19.0s	91%
16	Mistral Small Creative	97.3%	$0.0007	9.1s	92%
17	GPT-5.4 Nano (Reasoning, Low)	97.7%	$0.0055	20.6s	92%
18	GPT-5.4 Nano (Reasoning)	98.0%	$0.0061	24.5s	91%
19	Grok 4.20	98.5%	$0.0093	45.7s	92%
20	GPT-5.4 Mini (Reasoning, Low)	97.9%	$0.015	16.8s	92%
21	Hermes 3 405B	97.9%	$0.0032	53.2s	93%
22	GPT-4o, May 13th (temp=1)	98.2%	$0.033	14.4s	93%
23	Grok 4.20 (Beta)	98.1%	$0.018	15.8s	91%
24	GPT-5.4 Mini	97.8%	$0.015	16.8s	91%
25	Ministral 3 3B	97.3%	$0.0005	11.1s	90%
26	DeepSeek V3 (2024-12-26)	97.8%	$0.0021	54.6s	92%
27	Qwen 3.6 Flash	98.3%	$0.010	41.4s	91%
28	Stealth: Aurora Alpha	97.0%	$0.0000	9.8s	90%
29	GPT-5.4 Nano	97.3%	$0.0057	26.3s	91%
30	GPT-4o Mini (temp=0)	97.4%	$0.0012	34.8s	91%
31	Qwen 3 32B	97.6%	$0.0015	54.6s	92%
32	GPT-4.1 Nano	97.0%	$0.0007	13.3s	90%
33	Qwen 3.5 Plus (2026-02-15)	97.6%	$0.0060	31.5s	90%
34	Qwen 3.5 Flash	97.5%	$0.0025	47.5s	91%
35	Mistral Medium 3.1	97.5%	$0.0048	36.5s	91%
36	Gemini 2.5 Flash	97.3%	$0.0052	10.6s	89%
37	DeepSeek-V2 Chat	97.5%	$0.0021	53.3s	91%
38	Gemma 3 4B	96.7%	$0.0002	20.0s	90%
39	Gemma 3 12B	97.1%	$0.0004	41.3s	91%
40	Gemma 3 27B	97.2%	$0.0006	52.6s	91%
41	GPT-4o, Aug. 6th (temp=1)	97.9%	$0.018	24.4s	90%
42	Qwen 3.6 35B	98.2%	$0.0083	1.0m	91%
43	Qwen 3.5 9B	97.9%	$0.0011	1.4m	92%
44	Qwen 3.5 122B	98.1%	$0.025	1.1m	93%
45	Ministral 3B	96.5%	$0.0001	8.1s	89%
46	Claude Sonnet 4.5	98.1%	$0.035	38.1s	92%
47	ByteDance Seed 1.6 Flash	96.6%	$0.0013	27.3s	90%
48	Ministral 3 14B	96.5%	$0.0007	11.7s	89%
49	GPT-5.1	99.5%	$0.054	1.8m	96%
50	Ministral 3 8B	96.3%	$0.0008	19.6s	90%
51	Claude Haiku 4.5	96.9%	$0.011	21.6s	90%
52	Claude 3.7 Sonnet	98.1%	$0.042	46.7s	93%
53	Grok 4.20 (Beta, Reasoning)	98.0%	$0.039	34.0s	92%
54	Inception Mercury 2	96.8%	$0.0032	7.0s	87%
55	Z.AI GLM 4.5	97.0%	$0.0051	42.1s	90%
56	DeepSeek V4 Flash (Reasoning)	96.6%	$0.0007	31.1s	89%
57	GPT-5.4	98.8%	$0.049	1.4m	95%
58	Z.AI GLM 5 Turbo	97.0%	$0.0081	33.2s	89%
59	Mistral Large	96.8%	$0.014	30.9s	90%
60	Mistral NeMO	95.9%	$0.0005	10.1s	88%
61	Z.AI GLM 5	97.1%	$0.0084	1.2m	91%
62	GPT-5.4 Mini (Reasoning)	97.5%	$0.022	28.1s	89%
63	Grok 4	99.1%	$0.048	1.7m	94%
64	Qwen 3.5 Plus (2026-04-20)	98.3%	$0.017	1.8m	92%
65	Grok 4.20 (Reasoning)	98.4%	$0.018	1.5m	90%
66	Qwen 3.5 27B	97.8%	$0.020	1.6m	92%
67	Gemini 2.5 Flash Lite (Reasoning)	96.2%	$0.0028	30.8s	89%
68	GPT-5.2	98.7%	$0.056	1.5m	94%
69	Claude 3 Haiku	95.9%	$0.0025	14.9s	88%
70	MiniMax M2.5	96.5%	$0.0034	1.3m	91%
71	Arcee AI: Trinity Large (Preview)	96.1%	$0.0000	43.6s	89%
72	MiniMax M2.7	96.7%	$0.0040	1.1m	90%
73	DeepSeek V4 Flash	96.0%	$0.0006	31.6s	88%
74	GPT-5 Nano	96.7%	$0.0042	1.4m	91%
75	Z.AI GLM 5.1	97.3%	$0.014	1.5m	91%
76	Gemini 2.5 Flash Lite	95.4%	$0.0009	9.5s	87%
77	Ministral 8B	95.6%	$0.0004	10.4s	86%
78	GPT-5.4 (Reasoning, Low)	98.4%	$0.055	1.4m	93%
79	Z.AI GLM 4.5 Air	96.2%	$0.0029	58.2s	88%
80	GPT-4o, May 13th (temp=0)	96.1%	$0.035	14.1s	89%
81	Mistral Large 3	95.5%	$0.0033	30.3s	87%
82	Claude 3.5 Sonnet	97.1%	$0.048	35.4s	90%
83	GPT-OSS 120B	97.0%	$0.0015	1.8m	88%
84	Rocinante 12B	95.9%	$0.0014	38.4s	86%
85	Claude Sonnet 4	96.6%	$0.032	43.7s	89%
86	Mistral Large 2	96.2%	$0.013	29.4s	86%
87	Arcee AI: Trinity Mini	95.3%	$0.0003	9.2s	85%
88	Stealth: Hunter Alpha	95.8%	$0.0000	55.0s	87%
89	Gemini 3.1 Flash Lite (Reasoning)	95.0%	$0.0030	11.9s	86%
90	Xiaomi MIMO v2.5 Pro	95.8%	$0.0085	53.5s	87%
91	DeepSeek V4 Pro	96.7%	$0.0048	1.3m	87%
92	Gemini 3 Pro (Preview)	96.9%	$0.055	54.4s	91%
93	Stealth: Healer Alpha	94.9%	$0.0000	23.7s	85%
94	Z.AI GLM 4.7	96.0%	$0.010	1.4m	88%
95	Qwen 3.5 397B A17B	98.2%	$0.014	3.0m	91%
96	Gemini 3.1 Flash Lite (Preview)	94.5%	$0.0030	8.4s	84%
97	Gemini 3.1 Flash Lite	94.6%	$0.0030	12.1s	84%
98	Hermes 3 70B	95.5%	$0.0010	1.2m	86%
99	Grok 4.3 (Reasoning)	98.0%	$0.021	2.3m	88%
100	GPT-4o, Aug. 6th (temp=0)	95.6%	$0.023	22.7s	85%
101	GPT-5	98.8%	$0.065	2.8m	95%
102	Qwen 3.6 27B	97.6%	$0.025	2.3m	89%
103	GPT-5 Mini	95.3%	$0.0100	57.4s	86%
104	Xiaomi MIMO v2.5	94.7%	$0.0054	31.8s	85%
105	Qwen3.6 Max Preview	99.1%	$0.050	3.5m	94%
106	Claude Opus 4.5	97.1%	$0.070	53.4s	90%
107	Cohere Command R+ (Aug. 2024)	95.5%	$0.020	52.5s	86%
108	Llama 3.1 8B	95.4%	$0.0003	1.3m	85%
109	Gemini 3 Flash (Preview)	94.1%	$0.0078	19.6s	85%
110	GPT-5.5 (Reasoning)	99.6%	$0.142	1.8m	97%
111	Qwen 3.5 35B	96.1%	$0.018	1.0m	84%
112	Gemma 4 26B	94.9%	$0.0009	55.1s	84%
113	GPT-5.4 (Reasoning)	98.9%	$0.089	2.6m	95%
114	MoonshotAI: Kimi K2.5	97.9%	$0.019	3.2m	89%
115	Z.AI GLM 4.7 Flash	94.5%	$0.0017	1.2m	85%
116	GPT-5.5 (Reasoning, Low)	99.3%	$0.139	1.8m	96%
117	DeepSeek V3.2	95.3%	$0.0014	1.9m	86%
118	Gemini 3 Flash (Preview, Reasoning)	94.4%	$0.012	30.1s	83%
119	GPT-5.5	99.2%	$0.139	1.7m	95%
120	Gemma 4 31B (Reasoning)	95.5%	$0.0014	2.2m	87%
121	Nemotron 3 Super	95.2%	$0.0000	1.4m	83%
122	Gemma 4 31B	94.9%	$0.0010	1.6m	85%
123	Gemma 4 26B (Reasoning)	95.0%	$0.0013	2.0m	85%
124	Gemini 3.1 Pro (Preview)	98.3%	$0.107	1.8m	91%
125	Aion 2.0	94.3%	$0.0064	1.3m	84%
126	Z.AI GLM 4.6	93.7%	$0.0065	51.5s	83%
127	Claude Opus 4.6	96.4%	$0.078	1.2m	88%
128	Qwen 2.5 72B	93.3%	$0.0010	36.7s	81%
129	Gemini 2.5 Pro	94.1%	$0.036	36.2s	84%
130	DeepSeek V4 Pro (Reasoning)	96.1%	$0.015	3.1m	87%
131	Nemotron 3 Nano	94.4%	$0.0010	1.1m	80%
132	Llama 3.1 Nemotron 70B	93.2%	$0.0038	31.7s	80%
133	Claude Sonnet 4.6	93.7%	$0.031	39.3s	83%
134	Claude Opus 4.6 (Reasoning)	96.2%	$0.088	1.4m	88%
135	WizardLM 2 8x22b	94.8%	$0.0026	1.8m	80%
136	Claude Opus 4.7 (Reasoning)	94.4%	$0.076	32.0s	85%
137	Llama 3.1 70B	92.5%	$0.0015	29.4s	78%
138	Claude Sonnet 4.6 (Reasoning)	94.6%	$0.060	1.2m	84%
139	DeepSeek V3.1	93.2%	$0.0020	1.8m	81%
140	Claude Opus 4.7	93.8%	$0.069	30.4s	83%
141	Claude Opus 4	98.5%	$0.209	1.4m	94%
142	MoonshotAI: Kimi K2.6	98.5%	$0.058	6.5m	92%
143	Mistral Small 3.2 24B	95.8%	$0.0069	5.7m	84%
144	Inception Mercury	90.1%	$0.011	17.6s	71%
145	ByteDance Seed 1.6	92.1%	$0.013	2.5m	76%
146	ByteDance Seed 2.0 Lite	89.5%	$0.012	2.2m	71%
147	ByteDance Seed 2.0 Mini	87.5%	$0.0045	4.9m	70%
96.66%

Individual Scenarios

Detailed Writing Rules

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 5.1	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-5.5 (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100.0%
GPT-5	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%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Qwen 3.6 27B	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 M2.7	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100.0%
Qwen 3.6 35B	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%
GPT-OSS 120B	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Qwen 3.5 9B	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%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
Z.AI GLM 4.5 Air	100	100	100	100	100	100.0%
DeepSeek V4 Pro	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	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 V4 Flash	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
Grok 4.20	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Grok 4.3	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Qwen3.6 Max Preview	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	99.9%
Writer: Palmyra X5	100	100	100	100	99	99.9%
GPT-4.1 Nano	100	100	100	100	99	99.9%
Grok 4.3 (Reasoning)	100	100	100	100	99	99.9%
GPT-5.4 (Reasoning, Low)	100	100	100	100	99	99.9%
Qwen 3.6 Flash	100	100	100	100	99	99.9%
Grok 4 Fast	100	100	100	100	99	99.9%
Qwen 3.5 27B	100	100	100	100	99	99.8%
GPT-5.4	100	100	100	100	99	99.8%
Grok 4.20 (Reasoning)	100	100	100	100	99	99.8%
Claude Sonnet 4.5	100	100	100	100	99	99.7%
Inception Mercury 2	100	100	100	100	99	99.7%
Claude Opus 4.6	100	100	100	100	98	99.7%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	98	99.6%
GPT-5.4 Nano	100	100	100	100	98	99.6%
Claude Sonnet 4	100	100	100	99	99	99.6%
LFM2 24B	100	100	100	100	98	99.6%
Claude Haiku 4.5	100	100	100	100	98	99.5%
Claude 3.7 Sonnet	100	100	100	100	97	99.5%
Gemini 2.5 Flash	100	100	100	100	97	99.5%
Arcee AI: Trinity Mini	100	100	100	99	98	99.5%
Xiaomi MIMO v2.5 Pro	100	100	100	99	98	99.4%
Z.AI GLM 4.7	100	100	100	99	98	99.4%
Claude Opus 4.6 (Reasoning)	100	100	100	99	98	99.4%
Qwen 3.5 Plus (2026-04-20)	100	100	100	100	97	99.4%
GPT-5.4 Mini (Reasoning)	100	100	100	100	97	99.3%
Mistral Small 4	100	100	100	100	97	99.3%
Mistral Medium 3.1	100	100	100	100	96	99.3%
Claude Sonnet 4.6 (Reasoning)	100	100	100	98	98	99.3%
Qwen 3.5 35B	100	100	100	100	96	99.3%
Qwen 3.5 397B A17B	100	100	100	98	98	99.3%
Rocinante 12B	100	100	100	99	98	99.2%
GPT-4o, May 13th (temp=1)	100	100	100	100	96	99.2%
Claude Opus 4.7	100	100	100	100	96	99.2%
GPT-5.4 Nano (Reasoning)	100	100	100	100	96	99.2%
MoonshotAI: Kimi K2.5	100	100	100	100	96	99.2%
DeepSeek V4 Pro (Reasoning)	100	100	100	100	96	99.1%
GPT-4.1 Mini	100	100	100	100	95	99.1%
GPT-5 Nano	100	100	100	99	96	99.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	98	97	99.0%
Ministral 8B	100	100	100	100	95	99.0%
Ministral 3 14B	100	100	100	99	96	99.0%
Qwen 3.5 Flash	100	100	100	100	95	98.9%
Stealth: Aurora Alpha	100	100	100	100	94	98.9%
Z.AI GLM 5	100	100	100	100	94	98.8%
Gemma 3 27B	100	100	100	100	94	98.7%
Qwen 3 32B	100	100	100	99	94	98.7%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	93	98.6%
Nemotron 3 Super	100	100	100	98	95	98.6%
GPT-5 Mini	100	100	99	99	95	98.6%
Gemma 3 4B	100	100	100	96	96	98.5%
Llama 3.1 8B	100	100	100	97	95	98.5%
Mistral Large	100	100	100	99	93	98.5%
Gemma 4 31B	100	100	99	97	96	98.5%
Claude Opus 4.7 (Reasoning)	100	100	100	97	95	98.4%
Aion 2.0	99	99	98	98	98	98.3%
Hermes 3 405B	100	100	100	98	93	98.3%
Xiaomi MIMO v2.5	100	100	98	98	95	98.3%
Stealth: Hunter Alpha	100	100	100	97	94	98.2%
Ministral 3B	100	100	100	96	95	98.2%
GPT-4o, Aug. 6th (temp=0)	100	100	100	99	92	98.2%
Ministral 3 8B	100	100	100	96	94	98.1%
WizardLM 2 8x22b	100	100	100	98	92	98.0%
Gemma 4 26B (Reasoning)	100	100	100	96	94	98.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	99	98	92	97.8%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	98	91	97.7%
MiniMax M2.5	100	100	100	95	93	97.6%
ByteDance Seed 1.6	100	100	100	98	90	97.5%
Gemma 3 12B	100	100	100	94	93	97.5%
Gemini 3 Pro (Preview)	100	100	100	94	93	97.5%
Gemini 3.1 Flash Lite (Preview)	100	100	100	97	90	97.4%
DeepSeek V3.2	100	99	98	97	93	97.4%
DeepSeek V4 Flash (Reasoning)	100	100	98	97	92	97.3%
Stealth: Healer Alpha	100	100	100	98	89	97.3%
Claude 3.5 Sonnet	100	100	100	98	89	97.3%
Mistral Small Creative	100	100	100	99	88	97.2%
Gemma 4 31B (Reasoning)	100	100	100	94	92	97.1%
Mistral Large 3	100	99	96	95	93	96.8%
Gemini 3.1 Flash Lite	100	100	97	96	91	96.6%
Claude Opus 4	100	97	96	95	95	96.6%
Nemotron 3 Nano	100	100	96	95	92	96.5%
Arcee AI: Trinity Large (Preview)	100	100	100	95	87	96.5%
Gemini 2.5 Pro	100	100	96	94	93	96.5%
ByteDance Seed 1.6 Flash	100	98	96	95	93	96.4%
Gemini 3 Flash (Preview)	100	100	98	95	90	96.4%
GPT-4o, May 13th (temp=0)	100	100	100	100	81	96.2%
Hermes 3 70B	100	100	95	94	91	96.1%
Gemini 3.1 Flash Lite (Reasoning)	100	100	100	97	83	96.0%
Ministral 3 3B	100	100	100	96	84	96.0%
Inception Mercury	100	100	100	100	79	95.8%
Gemma 4 26B	99	97	97	94	91	95.7%
Mistral Large 2	100	99	96	96	88	95.6%
DeepSeek V3.1	100	96	96	95	91	95.6%
Z.AI GLM 4.6	100	100	96	93	88	95.6%
Claude Sonnet 4.6	100	100	96	94	84	94.9%
Claude 3 Haiku	100	100	97	91	85	94.7%
Llama 3.1 Nemotron 70B	100	100	96	94	83	94.6%
Cohere Command R+ (Aug. 2024)	100	100	100	88	84	94.4%
Gemini 2.5 Flash Lite	100	96	93	91	89	94.0%
Mistral NeMO	100	99	98	95	78	93.9%
Z.AI GLM 4.7 Flash	100	100	92	91	84	93.5%
Qwen 2.5 72B	100	100	99	96	71	93.1%
Mistral Small 3.2 24B	100	97	92	89	87	92.8%
ByteDance Seed 2.0 Mini	100	100	99	87	78	92.7%
Llama 3.1 70B	100	97	96	83	70	89.2%
ByteDance Seed 2.0 Lite	98	96	79	73	72	83.9%

▼

Horror: alone in an eerie place at night

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen3.6 Max Preview	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-04-20)	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%
Grok 4.20 (Reasoning)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.6 Flash	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%
Qwen 3.6 27B	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100.0%
Qwen 3.6 35B	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 Opus 4	100	100	100	100	100	100.0%
Xiaomi MIMO v2.5 Pro	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	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Grok 4.3	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	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%
Z.AI GLM 5 Turbo	100	100	100	100	100	100.0%
Ministral 3 14B	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	99.9%
Writer: Palmyra X5	100	100	100	100	100	99.9%
Grok 4.20	100	100	100	100	100	99.9%
MiniMax M2.7	100	100	100	100	100	99.9%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	99.9%
Qwen 3.5 397B A17B	100	100	100	100	99	99.8%
Z.AI GLM 5	100	100	100	100	99	99.8%
Grok 4.3 (Reasoning)	100	100	100	100	99	99.8%
Claude Sonnet 4.5	100	100	100	100	99	99.8%
GPT-5.4 (Reasoning)	100	100	100	100	99	99.8%
DeepSeek V3 (2025-03-24)	100	100	100	100	99	99.8%
Ministral 3 3B	100	100	100	100	99	99.8%
Qwen 3.5 122B	100	100	100	100	99	99.8%
Mistral Small 4 (Reasoning)	100	100	100	100	99	99.7%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	98	99.7%
Qwen3 235B A22B Instruct 2507	100	100	100	99	99	99.7%
Mistral Large 3	100	100	100	100	98	99.6%
GPT-5 Mini	100	100	100	100	99	99.6%
Qwen 3.5 35B	100	100	100	100	98	99.6%
Qwen 3.5 Flash	100	100	100	100	98	99.6%
MiniMax M2.5	100	100	100	100	99	99.6%
Mistral Large 2	100	100	100	100	98	99.6%
Z.AI GLM 5.1	100	100	100	100	98	99.6%
Claude Opus 4.6	100	100	100	100	98	99.6%
DeepSeek V4 Pro	100	100	100	100	98	99.5%
GPT-4.1 Mini	100	100	100	100	97	99.4%
Claude 3.5 Sonnet	100	100	100	100	97	99.4%
GPT-4o, May 13th (temp=1)	100	100	100	99	98	99.4%
Qwen 3.5 9B	100	100	100	100	97	99.4%
DeepSeek V4 Pro (Reasoning)	100	100	99	99	98	99.4%
Claude Opus 4.5	100	100	100	98	98	99.3%
Nemotron 3 Super	100	100	100	100	96	99.3%
Qwen 3.5 27B	100	100	100	100	96	99.3%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	96	99.3%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	96	99.2%
Hermes 3 405B	100	100	100	100	96	99.2%
Gemini 2.5 Flash	100	100	100	98	97	99.1%
Mistral Small Creative	100	100	100	97	97	98.9%
Rocinante 12B	100	100	100	98	97	98.9%
Gemma 3 12B	100	100	99	99	96	98.9%
Claude Haiku 4.5	100	100	100	97	96	98.7%
GPT-4o Mini (temp=1)	100	100	100	100	93	98.7%
Gemma 4 26B (Reasoning)	100	100	98	98	97	98.5%
Claude Opus 4.6 (Reasoning)	100	100	99	97	96	98.3%
Gemma 3 27B	100	99	99	97	96	98.3%
Stealth: Hunter Alpha	100	100	99	98	95	98.2%
Arcee AI: Trinity Large (Preview)	100	100	100	100	91	98.1%
DeepSeek V3 (2024-12-26)	100	100	100	95	95	98.1%
DeepSeek V3.2	100	100	99	97	95	98.0%
Qwen 3 32B	100	100	99	96	94	98.0%
Arcee AI: Trinity Mini	100	100	100	100	90	97.9%
Mistral Medium 3.1	100	100	99	98	93	97.9%
GPT-4o Mini (temp=0)	100	100	100	100	89	97.7%
Claude Sonnet 4.6 (Reasoning)	100	99	98	95	95	97.5%
DeepSeek V4 Flash (Reasoning)	100	100	100	96	92	97.5%
Inception Mercury 2	100	100	99	97	92	97.5%
DeepSeek-V2 Chat	100	100	100	100	88	97.5%
Mistral Large	100	100	100	96	92	97.5%
Gemini 3.1 Flash Lite	100	100	97	97	93	97.5%
Gemini 3.1 Flash Lite (Reasoning)	100	100	100	98	89	97.4%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	94	93	97.3%
DeepSeek V4 Flash	100	100	100	94	93	97.3%
Gemini 3 Pro (Preview)	100	100	98	98	91	97.3%
WizardLM 2 8x22b	100	97	97	96	95	97.3%
Aion 2.0	100	100	98	94	93	97.1%
Stealth: Aurora Alpha	100	100	98	94	93	97.0%
Gemma 4 31B (Reasoning)	100	100	100	94	90	96.9%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	85	96.9%
Z.AI GLM 4.5	100	100	96	95	92	96.7%
Cohere Command R+ (Aug. 2024)	100	100	96	93	93	96.5%
GPT-OSS 120B	100	100	100	100	82	96.5%
GPT-4o, May 13th (temp=0)	100	100	99	93	90	96.4%
Gemini 2.5 Pro	100	97	96	96	93	96.3%
Z.AI GLM 4.5 Air	100	100	100	95	86	96.2%
Z.AI GLM 4.7	100	99	98	96	87	96.1%
GPT-5 Nano	100	99	98	94	90	96.1%
Ministral 3 8B	100	100	95	93	92	96.0%
Llama 3.1 70B	100	100	96	93	90	95.9%
Stealth: Healer Alpha	98	98	97	96	91	95.8%
ByteDance Seed 1.6 Flash	100	98	95	93	92	95.8%
Gemma 4 31B	100	100	94	93	92	95.8%
Z.AI GLM 4.7 Flash	100	99	97	93	89	95.6%
Claude Opus 4.7	100	99	99	93	88	95.6%
Gemini 3 Flash (Preview)	100	100	96	93	89	95.6%
Ministral 3B	100	100	100	100	78	95.5%
Nemotron 3 Nano	100	100	98	91	89	95.4%
Gemini 3.1 Flash Lite (Preview)	100	100	98	96	82	95.4%
Ministral 8B	100	99	97	95	86	95.1%
Gemma 3 4B	100	98	97	91	90	95.1%
Claude 3 Haiku	100	100	99	94	83	95.1%
Gemma 4 26B	100	97	97	96	83	94.6%
Claude Sonnet 4.6	100	95	94	94	90	94.3%
Claude Opus 4.7 (Reasoning)	100	99	94	93	85	94.2%
Hermes 3 70B	100	100	100	100	70	94.0%
GPT-4o, Aug. 6th (temp=0)	100	100	96	90	83	93.9%
GPT-4.1 Nano	100	100	97	97	75	93.9%
Mistral NeMO	100	99	93	89	88	93.9%
Qwen 2.5 72B	100	100	93	90	86	93.8%
Z.AI GLM 4.6	97	96	95	93	86	93.4%
DeepSeek V3.1	100	100	98	97	72	93.3%
Xiaomi MIMO v2.5	100	97	95	93	83	93.3%
ByteDance Seed 1.6	100	100	94	90	80	92.6%
Llama 3.1 8B	100	100	99	93	67	91.9%
Gemini 2.5 Flash Lite	100	98	91	91	79	91.8%
ByteDance Seed 2.0 Lite	98	97	95	86	79	91.1%
LFM2 24B	96	95	92	87	85	91.1%
ByteDance Seed 2.0 Mini	100	94	89	87	81	90.1%
Llama 3.1 Nemotron 70B	97	92	92	85	76	88.5%
Inception Mercury	100	98	94	84	66	88.5%

▼

Literary fiction: old friends reunite

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100.0%
Qwen3.6 Max Preview	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-5.5 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	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 Plus (2026-04-20)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	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%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100.0%
Qwen 3.6 35B	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%
GPT-OSS 120B	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%
GPT-5.4 Mini (Reasoning, Low)	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%
Grok 4.20	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Qwen 3.6 Flash	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100.0%
Grok 4.20 (Reasoning)	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	99.9%
Mistral NeMO	100	100	100	100	100	99.9%
GPT-5.4	100	100	100	100	99	99.9%
GPT-4o, May 13th (temp=1)	100	100	100	100	99	99.9%
Grok 4.3 (Reasoning)	100	100	100	100	99	99.8%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	99	99.8%
Mistral Small 4	100	100	100	100	99	99.8%
Qwen 3 32B	100	100	100	100	99	99.8%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	99	99.8%
GPT-5.4 Nano	100	100	100	99	99	99.8%
DeepSeek-V2 Chat	100	100	100	100	98	99.6%
Writer: Palmyra X5	100	100	100	100	98	99.6%
Z.AI GLM 4.7	100	100	100	99	99	99.6%
Mistral Large 2	100	100	100	100	98	99.6%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	98	99.6%
Z.AI GLM 5	100	100	100	99	99	99.6%
Stealth: Hunter Alpha	100	100	100	100	98	99.6%
Gemini 2.5 Flash	100	100	100	100	98	99.5%
Qwen 3.5 Flash	100	100	100	100	98	99.5%
Gemma 4 31B (Reasoning)	100	100	100	100	98	99.5%
DeepSeek V4 Flash (Reasoning)	100	100	100	100	98	99.4%
GPT-5.4 Nano (Reasoning)	100	100	100	100	97	99.4%
DeepSeek V3.2	100	99	99	99	99	99.4%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	98	98	99.3%
Qwen 3.5 Plus (2026-02-15)	100	100	100	98	98	99.3%
GPT-5.4 Mini	100	100	100	100	96	99.2%
Mistral Small Creative	100	100	100	100	96	99.2%
Qwen 3.6 27B	100	100	100	100	96	99.2%
Gemini 3 Pro (Preview)	100	100	100	99	97	99.1%
Ministral 3 14B	100	100	100	100	96	99.1%
Claude Opus 4.5	100	100	100	100	96	99.1%
Z.AI GLM 5 Turbo	100	100	100	100	96	99.1%
GPT-4.1 Mini	100	100	100	98	98	99.1%
GPT-5.2	100	100	100	100	95	99.1%
Claude 3.5 Sonnet	100	100	100	98	97	99.1%
GPT-4o Mini (temp=1)	100	100	100	100	96	99.1%
MoonshotAI: Kimi K2.5	100	100	100	100	95	99.0%
Nemotron 3 Super	100	100	100	99	96	99.0%
Gemma 4 31B	100	100	100	100	95	99.0%
Nemotron 3 Nano	100	100	100	100	95	99.0%
Ministral 3 3B	100	100	99	98	97	98.9%
DeepSeek V3 (2024-12-26)	100	100	100	99	95	98.9%
MiniMax M2.5	100	100	100	100	94	98.9%
GPT-4.1 Nano	100	100	100	99	95	98.9%
Grok 4.3	100	100	100	100	94	98.9%
DeepSeek V4 Pro (Reasoning)	100	100	100	98	96	98.8%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	94	98.7%
Aion 2.0	100	100	100	100	93	98.6%
DeepSeek V3 (2025-03-24)	100	100	100	98	95	98.6%
LFM2 24B	100	100	100	96	96	98.6%
Z.AI GLM 4.5 Air	100	100	100	100	93	98.5%
MiniMax M2.7	100	100	100	97	96	98.5%
Z.AI GLM 5.1	100	100	100	100	93	98.5%
Gemma 3 27B	100	100	100	98	94	98.5%
WizardLM 2 8x22b	100	100	100	99	93	98.5%
ByteDance Seed 1.6 Flash	100	100	98	98	96	98.4%
Hermes 3 405B	100	100	100	100	92	98.4%
Claude Sonnet 4.5	100	100	100	96	95	98.2%
GPT-5 Mini	100	100	100	96	94	98.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	90	98.0%
Mistral Large	100	100	100	95	94	97.9%
Gemini 2.5 Flash Lite	100	100	100	96	93	97.9%
Ministral 3 8B	100	100	99	96	94	97.9%
Qwen 2.5 72B	100	100	100	100	90	97.9%
Claude Sonnet 4	100	100	100	96	93	97.7%
Ministral 8B	100	100	100	95	94	97.7%
Arcee AI: Trinity Mini	100	100	100	100	89	97.7%
Hermes 3 70B	100	100	100	96	93	97.7%
Claude Opus 4.6	100	99	97	97	95	97.6%
GPT-5 Nano	100	100	100	97	90	97.5%
Qwen 3.5 9B	100	100	99	96	92	97.4%
Claude 3 Haiku	100	100	99	95	93	97.4%
Xiaomi MIMO v2.5 Pro	100	100	100	100	87	97.4%
Gemini 3.1 Flash Lite	100	100	99	94	94	97.4%
Gemini 3.1 Flash Lite (Preview)	100	100	100	94	93	97.4%
DeepSeek V4 Flash	100	100	100	99	87	97.3%
Gemini 2.5 Flash Lite (Reasoning)	100	99	98	96	93	97.2%
Claude Haiku 4.5	100	100	98	94	94	97.2%
DeepSeek V4 Pro	100	100	100	93	93	97.1%
Gemini 2.5 Pro	100	100	100	97	87	96.8%
Stealth: Healer Alpha	100	99	95	95	95	96.6%
Gemini 3.1 Flash Lite (Reasoning)	100	100	100	100	83	96.6%
Z.AI GLM 4.7 Flash	100	99	97	95	92	96.6%
Mistral Medium 3.1	100	100	100	100	83	96.6%
Llama 3.1 8B	100	100	100	94	89	96.5%
ByteDance Seed 1.6	100	100	96	96	91	96.5%
Gemma 4 26B (Reasoning)	100	100	99	93	90	96.4%
Xiaomi MIMO v2.5	100	100	99	96	87	96.3%
Gemma 4 26B	100	100	96	92	92	96.0%
Llama 3.1 Nemotron 70B	100	100	96	96	89	96.0%
ByteDance Seed 2.0 Lite	100	100	96	96	87	95.9%
Mistral Large 3	100	100	100	90	88	95.7%
ByteDance Seed 2.0 Mini	100	99	98	91	89	95.3%
Mistral Small 3.2 24B	100	100	100	100	76	95.2%
Cohere Command R+ (Aug. 2024)	100	100	96	93	86	95.1%
Gemma 3 12B	100	97	96	91	90	94.9%
Z.AI GLM 4.6	100	100	93	92	90	94.8%
Claude Sonnet 4.6 (Reasoning)	97	94	94	93	93	94.4%
Claude Opus 4.7	100	100	100	91	80	94.2%
Claude Opus 4.7 (Reasoning)	100	100	99	91	78	93.5%
DeepSeek V3.1	100	99	96	86	85	93.3%
Gemma 3 4B	100	95	92	91	87	93.1%
Ministral 3B	100	100	88	88	86	92.5%
Gemini 3 Flash (Preview)	100	99	94	85	82	91.9%
Inception Mercury	100	100	100	83	76	91.9%
Claude Sonnet 4.6	97	93	91	90	88	91.9%
Rocinante 12B	100	100	95	93	55	88.7%
Llama 3.1 70B	100	89	88	80	62	83.7%

▼

Mystery: examining a crime scene

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen3.6 Max Preview	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-5.5 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-04-20)	100	100	100	100	100	100.0%
Qwen 3.6 Flash	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100.0%
Qwen 3.6 35B	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%
Grok 4 Fast	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100.0%
Grok 4.3	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	99.9%
Inception Mercury 2	100	100	100	100	99	99.9%
Grok 4.20	100	100	100	100	99	99.8%
GPT-5.5 (Reasoning, Low)	100	100	100	100	99	99.8%
Grok 4.20 (Beta)	100	100	100	100	99	99.7%
o4 Mini High	100	100	100	100	98	99.7%
Qwen 3.6 27B	100	100	100	100	98	99.6%
Qwen 3.5 397B A17B	100	100	100	99	98	99.6%
GPT-5.4 Mini (Reasoning)	100	100	100	99	99	99.5%
GPT-5.4 Mini	100	100	100	99	98	99.4%
Writer: Palmyra X5	100	100	100	100	97	99.4%
Qwen 3.5 27B	100	100	100	99	97	99.2%
GPT-5.4 Nano (Reasoning)	100	100	100	99	97	99.2%
MoonshotAI: Kimi K2.5	100	100	100	100	96	99.2%
MiniMax M2.7	100	100	100	100	96	99.1%
GPT-OSS 120B	100	100	100	98	97	99.1%
Claude 3 Haiku	100	100	100	99	96	99.1%
Rocinante 12B	100	100	100	100	95	99.0%
DeepSeek-V2 Chat	100	100	100	98	97	98.9%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	95	98.9%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	95	98.9%
Grok 4.20 (Reasoning)	100	100	100	99	95	98.8%
GPT-5	100	100	100	97	97	98.8%
Claude Opus 4	100	100	100	100	94	98.8%
Z.AI GLM 5	100	100	99	98	96	98.7%
Nemotron 3 Super	100	99	99	99	97	98.7%
GPT-5.4 (Reasoning, Low)	100	100	100	98	95	98.6%
Grok 4.3 (Reasoning)	100	100	100	98	94	98.5%
GPT-5.2	100	100	97	97	97	98.4%
Claude Sonnet 4.5	100	100	98	97	96	98.3%
Hermes 3 405B	100	100	100	98	93	98.2%
DeepSeek V3 (2025-03-24)	100	100	97	97	96	98.1%
Gemini 2.5 Flash (Reasoning)	100	100	99	98	93	98.1%
GPT-5.4 Nano (Reasoning, Low)	100	100	98	96	95	98.0%
Claude 3.5 Sonnet	100	100	100	99	91	97.9%
Gemini 2.5 Flash	100	98	97	97	97	97.9%
GPT-4o Mini (temp=1)	100	100	100	99	91	97.9%
Hermes 3 70B	100	100	97	96	96	97.9%
Llama 3.1 8B	100	100	99	97	93	97.8%
LFM2 24B	100	100	100	99	90	97.8%
GPT-4o Mini (temp=0)	100	100	99	96	94	97.7%
DeepSeek V4 Flash	100	100	99	99	89	97.5%
DeepSeek V4 Pro (Reasoning)	100	100	97	96	94	97.4%
Qwen 3.5 9B	100	100	100	100	86	97.2%
Claude Opus 4.6 (Reasoning)	100	100	97	95	93	97.1%
GPT-5.4	99	99	98	96	94	97.0%
Claude Haiku 4.5	100	99	97	96	93	97.0%
Qwen 3 32B	100	100	100	92	92	96.8%
Llama 3.1 Nemotron 70B	100	100	95	95	94	96.7%
Claude Sonnet 4	100	98	96	96	93	96.6%
DeepSeek V3 (2024-12-26)	100	100	100	100	83	96.6%
GPT-5.4 Mini (Reasoning, Low)	100	98	96	96	93	96.6%
Ministral 3B	100	100	100	99	84	96.5%
Xiaomi MIMO v2.5 Pro	100	100	97	95	90	96.4%
Nemotron 3 Nano	100	100	100	98	84	96.4%
Qwen 3.5 122B	100	100	100	91	89	96.1%
DeepSeek V4 Flash (Reasoning)	100	99	96	96	90	96.1%
Claude 3.7 Sonnet	100	99	99	94	88	96.1%
Claude Opus 4.5	100	100	100	93	87	96.0%
Gemini 3 Pro (Preview)	100	97	96	96	91	95.8%
ByteDance Seed 1.6 Flash	100	99	94	93	92	95.6%
Z.AI GLM 5.1	100	100	97	91	89	95.4%
GPT-5.4 Nano	100	99	97	91	90	95.4%
Gemma 4 31B (Reasoning)	100	100	96	92	88	95.2%
Ministral 3 3B	100	100	98	94	84	95.1%
Qwen 3.5 Flash	100	100	96	96	84	95.1%
Mistral NeMO	100	100	100	89	86	95.1%
Mistral Small 4 (Reasoning)	100	97	95	93	89	95.0%
Qwen 3.5 35B	98	96	96	94	91	95.0%
GPT-5 Nano	98	98	94	92	92	94.9%
Gemma 4 26B (Reasoning)	100	96	95	92	90	94.8%
Mistral Small Creative	98	97	96	94	88	94.7%
Gemma 3 27B	100	100	94	93	86	94.6%
GPT-5 Mini	100	98	98	96	81	94.6%
Mistral Small 4	100	100	97	92	84	94.6%
GPT-4o, May 13th (temp=1)	99	96	96	93	88	94.6%
Gemma 3 12B	100	96	95	94	87	94.4%
Claude Opus 4.6	100	97	96	90	89	94.4%
Claude Sonnet 4.6 (Reasoning)	100	98	94	92	87	94.3%
Mistral Small 3.2 24B	100	100	100	97	74	94.3%
Xiaomi MIMO v2.5	100	100	99	90	81	94.1%
Gemini 3.1 Flash Lite (Reasoning)	100	96	92	91	90	94.0%
Z.AI GLM 4.7 Flash	100	98	96	94	82	93.9%
Ministral 8B	99	97	96	91	87	93.8%
Z.AI GLM 4.5 Air	100	98	97	89	84	93.7%
Aion 2.0	100	99	94	88	88	93.6%
Cohere Command R+ (Aug. 2024)	100	100	100	98	70	93.6%
Gemma 4 31B	100	100	95	87	86	93.5%
Stealth: Healer Alpha	100	93	93	93	88	93.4%
GPT-4.1 Nano	100	97	90	90	90	93.3%
Arcee AI: Trinity Large (Preview)	100	100	92	91	84	93.1%
Z.AI GLM 4.5	100	100	90	90	85	93.0%
Ministral 3 8B	100	98	92	88	87	92.8%
Arcee AI: Trinity Mini	100	100	92	88	84	92.8%
Z.AI GLM 5 Turbo	100	100	94	94	75	92.7%
GPT-4o, May 13th (temp=0)	94	94	94	92	89	92.6%
GPT-4o, Aug. 6th (temp=0)	100	96	95	91	82	92.6%
Z.AI GLM 4.7	100	95	95	86	85	92.3%
Mistral Medium 3.1	96	95	94	88	86	91.9%
Gemma 3 4B	100	96	91	90	82	91.7%
DeepSeek V3.2	98	95	90	89	83	91.1%
MiniMax M2.5	96	95	93	88	82	90.9%
ByteDance Seed 1.6	100	90	89	89	87	90.9%
Ministral 3 14B	96	94	93	90	81	90.8%
Gemini 3.1 Flash Lite	100	99	94	81	78	90.4%
Gemini 3.1 Flash Lite (Preview)	97	93	91	86	84	90.3%
Llama 3.1 70B	100	99	96	94	61	89.9%
WizardLM 2 8x22b	95	94	92	84	84	89.8%
Gemini 3 Flash (Preview, Reasoning)	100	92	92	84	81	89.6%
Gemini 2.5 Pro	100	95	92	89	72	89.5%
DeepSeek V4 Pro	100	100	91	90	66	89.4%
Inception Mercury	100	100	100	83	62	89.1%
Claude Opus 4.7 (Reasoning)	100	90	87	84	83	89.0%
Mistral Large 2	100	99	98	83	65	88.8%
Gemini 2.5 Flash Lite	98	97	91	81	76	88.7%
Mistral Large 3	96	91	88	87	78	88.0%
DeepSeek V3.1	99	92	88	86	75	88.0%
Claude Opus 4.7	100	88	86	83	82	87.9%
Gemini 2.5 Flash Lite (Reasoning)	96	90	87	83	83	87.6%
Claude Sonnet 4.6	100	95	88	78	74	87.2%
ByteDance Seed 2.0 Mini	97	96	91	88	64	87.0%
Z.AI GLM 4.6	100	96	83	81	75	87.0%
Mistral Large	96	93	86	84	76	86.8%
Gemini 3 Flash (Preview)	95	89	85	85	79	86.5%
Gemma 4 26B	96	92	85	80	79	86.1%
Stealth: Hunter Alpha	98	87	84	84	70	84.8%
Qwen 2.5 72B	96	89	85	79	69	83.5%
ByteDance Seed 2.0 Lite	100	78	77	74	57	77.2%

▼

Romance: separated couple reunites

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen3.6 Max Preview	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 5.1	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100.0%
Grok 4.3 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	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%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-04-20)	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100.0%
Grok 4.20 (Reasoning)	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 3.6 Flash	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%
Qwen 3.6 27B	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%
GPT-5.5	100	100	100	100	100	100.0%
Qwen 3.6 35B	100	100	100	100	100	100.0%
DeepSeek V4 Flash (Reasoning)	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 Sonnet 4.5	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
GPT-OSS 120B	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%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	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-5.4 Mini	100	100	100	100	100	100.0%
DeepSeek V3.1	100	100	100	100	100	100.0%
Grok 4.20	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%
Grok 4.3	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%
Ministral 3 14B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	99.9%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	99.9%
GPT-5.4	100	100	100	100	100	99.9%
Z.AI GLM 4.5 Air	100	100	100	100	100	99.9%
Gemma 4 26B (Reasoning)	100	100	100	100	100	99.9%
Claude Sonnet 4	100	100	100	100	99	99.9%
Mistral Small 3.2 24B	100	100	100	100	99	99.9%
GPT-4o, May 13th (temp=1)	100	100	100	100	99	99.8%
Hermes 3 405B	100	100	100	100	99	99.8%
DeepSeek V4 Pro	100	100	100	100	99	99.8%
Claude Opus 4.6 (Reasoning)	100	100	100	100	99	99.8%
Z.AI GLM 4.5	100	100	100	100	99	99.8%
GPT-5.4 (Reasoning)	100	100	100	100	99	99.8%
DeepSeek-V2 Chat	100	100	100	100	99	99.8%
Qwen 3 32B	100	100	100	100	99	99.8%
Stealth: Hunter Alpha	100	100	100	100	99	99.8%
Claude Opus 4.5	100	100	100	100	99	99.7%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	99	99.7%
LFM2 24B	100	100	100	100	99	99.7%
Gemini 3.1 Flash Lite	100	100	100	100	99	99.7%
DeepSeek V3 (2025-03-24)	100	100	100	100	98	99.6%
Mistral NeMO	100	100	100	100	98	99.6%
Qwen3 235B A22B Instruct 2507	100	100	100	100	98	99.6%
Qwen 3.5 397B A17B	100	100	100	100	98	99.6%
DeepSeek V4 Pro (Reasoning)	100	100	100	100	98	99.6%
Arcee AI: Trinity Mini	100	100	100	99	99	99.6%
Gemini 2.5 Flash	100	100	100	100	97	99.5%
Gemma 4 31B (Reasoning)	100	100	100	99	98	99.5%
Qwen 3.5 27B	100	100	100	100	97	99.5%
Mistral Large 2	100	100	100	100	96	99.3%
Z.AI GLM 4.7	100	100	100	100	96	99.2%
Qwen 3.5 Flash	100	100	100	100	96	99.2%
Qwen 3.5 122B	100	100	100	100	96	99.2%
Mistral Medium 3.1	100	100	99	99	98	99.2%
GPT-5 Mini	100	100	100	99	97	99.2%
GPT-4.1 Mini	100	100	100	100	96	99.1%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	96	99.1%
Z.AI GLM 4.6	100	100	100	100	96	99.1%
Gemini 3 Flash (Preview)	100	100	100	99	96	99.1%
Mistral Small 4 (Reasoning)	100	100	100	100	95	99.1%
Mistral Large	100	100	100	99	96	99.0%
Z.AI GLM 4.7 Flash	100	100	99	98	98	99.0%
ByteDance Seed 1.6 Flash	100	100	100	100	95	99.0%
Mistral Small 4	100	100	100	100	95	99.0%
Xiaomi MIMO v2.5 Pro	100	100	100	99	96	98.9%
Gemma 3 12B	100	100	100	98	96	98.9%
Stealth: Aurora Alpha	100	100	100	98	96	98.8%
Gemma 3 4B	100	100	100	100	94	98.8%
GPT-4.1 Nano	100	100	100	99	95	98.7%
Gemma 4 31B	100	100	100	100	94	98.7%
Inception Mercury 2	100	100	100	100	93	98.7%
Claude Haiku 4.5	100	100	100	97	96	98.6%
Ministral 3B	100	100	100	98	95	98.5%
Gemini 2.5 Flash Lite	100	99	99	98	96	98.3%
Mistral Large 3	100	100	100	100	91	98.3%
GPT-5 Nano	100	100	99	96	96	98.2%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	97	93	98.2%
GPT-4o Mini (temp=0)	100	100	99	99	93	98.1%
DeepSeek V4 Flash	100	100	100	96	94	98.1%
Gemini 3 Pro (Preview)	100	100	98	97	93	97.8%
GPT-4o, May 13th (temp=0)	100	100	100	99	90	97.8%
Gemini 3.1 Flash Lite (Reasoning)	100	100	99	98	92	97.6%
Claude 3 Haiku	100	100	99	95	94	97.6%
Qwen 2.5 72B	100	100	100	95	93	97.6%
Claude Opus 4.7	100	100	100	96	92	97.5%
DeepSeek V3.2	99	98	98	97	96	97.5%
Xiaomi MIMO v2.5	100	100	99	97	91	97.4%
Ministral 3 3B	100	100	96	95	95	97.4%
Arcee AI: Trinity Large (Preview)	100	100	97	96	94	97.3%
ByteDance Seed 1.6	100	100	99	97	91	97.3%
Ministral 3 8B	100	100	100	100	86	97.2%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	85	97.1%
Cohere Command R+ (Aug. 2024)	100	100	100	100	85	96.9%
Gemini 3.1 Flash Lite (Preview)	100	98	98	95	94	96.8%
Gemma 3 27B	100	100	96	95	93	96.7%
Gemma 4 26B	100	100	98	93	92	96.7%
Ministral 8B	100	98	98	94	92	96.5%
Nemotron 3 Nano	100	100	100	96	86	96.4%
Stealth: Healer Alpha	100	100	97	94	89	96.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	99	79	95.7%
ByteDance Seed 2.0 Mini	100	99	97	92	88	95.3%
Claude Sonnet 4.6 (Reasoning)	100	100	98	93	83	94.9%
Claude Opus 4.7 (Reasoning)	100	99	95	91	90	94.9%
Llama 3.1 70B	100	100	96	92	86	94.8%
Llama 3.1 8B	100	100	96	89	89	94.7%
Rocinante 12B	100	100	98	88	86	94.2%
Claude Sonnet 4.6	100	100	96	94	82	94.2%
Gemini 2.5 Pro	100	96	92	90	90	93.7%
Hermes 3 70B	100	100	100	97	70	93.5%
Qwen 3.5 35B	100	100	100	100	39	87.8%
Llama 3.1 Nemotron 70B	100	99	97	93	49	87.5%
ByteDance Seed 2.0 Lite	100	100	97	52	50	79.9%
Inception Mercury	100	100	100	71	28	79.8%

▼

Thriller: chase through city streets

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen3.6 Max Preview	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-5.5 (Reasoning)	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-04-20)	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100.0%
Grok 4.20 (Reasoning)	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Qwen 3.6 Flash	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%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-5.5	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 Sonnet 4.5	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%
GPT-4.1 Mini	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
DeepSeek V4 Pro	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100.0%
Grok 4.20	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, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
Grok 4.3	100	100	100	100	100	99.9%
Stealth: Aurora Alpha	100	100	100	100	100	99.9%
Qwen 3.5 122B	100	100	100	100	100	99.9%
GPT-5.4	100	100	100	100	100	99.9%
GPT-4.1 Nano	100	100	100	100	100	99.9%
Grok 4.20 (Beta)	100	100	100	100	99	99.8%
Claude Opus 4	100	100	100	100	99	99.8%
Qwen 3.5 397B A17B	100	100	100	100	99	99.8%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	99	99.8%
Qwen 3.6 35B	100	100	100	100	99	99.7%
Inception Mercury 2	100	100	100	100	99	99.7%
DeepSeek V3 (2024-12-26)	100	100	100	100	98	99.6%
Claude Sonnet 4	100	100	100	100	98	99.6%
Mistral Small 4 (Reasoning)	100	100	100	99	99	99.6%
Qwen 3.6 27B	100	100	100	100	98	99.6%
GPT-5.2	100	100	100	100	97	99.3%
GPT-OSS 120B	100	100	100	99	98	99.3%
Gemini 2.5 Flash	100	100	100	100	97	99.3%
Mistral Small 3.2 24B	100	100	100	100	96	99.3%
Qwen 3.5 27B	100	100	100	99	98	99.3%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	96	99.3%
DeepSeek-V2 Chat	100	100	100	100	96	99.3%
LFM2 24B	100	100	100	99	98	99.3%
Qwen 3 32B	100	100	100	99	97	99.2%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	97	99.2%
Grok 4.20 (Beta, Reasoning)	100	100	100	98	98	99.1%
Claude Opus 4.6	100	100	100	99	97	99.1%
Claude Opus 4.5	100	100	100	98	97	99.1%
Gemma 4 26B	100	100	99	99	97	99.1%
Rocinante 12B	100	100	100	98	97	99.1%
Z.AI GLM 5.1	100	100	100	100	95	99.0%
Nemotron 3 Super	100	100	100	100	95	99.0%
Claude 3.5 Sonnet	100	100	100	99	96	99.0%
MiniMax M2.7	100	100	100	99	96	99.0%
Qwen 3.5 9B	100	100	100	98	97	99.0%
Grok 4.3 (Reasoning)	100	100	100	99	96	99.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	99	99	96	99.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	95	99.0%
GPT-4o Mini (temp=1)	100	100	100	100	95	98.9%
DeepSeek V3 (2025-03-24)	100	100	100	100	94	98.8%
Gemma 4 31B (Reasoning)	100	100	100	100	94	98.8%
Gemini 3 Pro (Preview)	100	100	98	98	97	98.6%
Ministral 3 14B	100	100	100	98	95	98.5%
DeepSeek V3.2	100	100	99	97	96	98.4%
Z.AI GLM 4.7	100	99	98	97	97	98.4%
MiniMax M2.5	100	100	100	97	95	98.3%
GPT-5.4 Nano	100	100	98	98	95	98.3%
GPT-5.4 Nano (Reasoning)	100	100	100	99	92	98.2%
Claude Opus 4.6 (Reasoning)	100	100	100	100	91	98.2%
Z.AI GLM 4.5 Air	100	100	100	99	92	98.2%
WizardLM 2 8x22b	100	100	100	99	92	98.2%
Z.AI GLM 5 Turbo	100	99	98	98	95	98.1%
Gemma 4 26B (Reasoning)	100	100	99	98	94	98.1%
Claude Sonnet 4.6 (Reasoning)	100	100	97	97	96	98.1%
Claude Opus 4.7 (Reasoning)	100	100	100	99	92	98.1%
Gemini 3 Flash (Preview, Reasoning)	100	100	99	98	93	98.1%
Mistral Medium 3.1	100	100	100	100	90	97.9%
GPT-5 Mini	100	100	99	97	94	97.9%
Z.AI GLM 4.5	100	100	99	98	93	97.9%
Llama 3.1 70B	100	100	100	99	91	97.8%
DeepSeek V4 Flash (Reasoning)	100	100	100	100	89	97.8%
Ministral 3 3B	100	100	100	100	89	97.8%
Gemma 3 12B	100	99	99	98	93	97.8%
Xiaomi MIMO v2.5	100	100	99	96	93	97.7%
ByteDance Seed 1.6 Flash	100	100	100	97	92	97.7%
GPT-4o Mini (temp=0)	100	100	100	97	91	97.6%
Cohere Command R+ (Aug. 2024)	100	100	99	99	89	97.5%
Claude 3 Haiku	100	100	100	97	90	97.4%
Qwen 3.5 35B	100	100	97	96	93	97.4%
Arcee AI: Trinity Large (Preview)	100	100	100	100	87	97.4%
GPT-4o, May 13th (temp=1)	100	100	98	97	92	97.3%
Gemma 3 4B	100	100	97	96	93	97.3%
DeepSeek V4 Flash	100	98	97	96	95	97.3%
Mistral Large	100	99	99	99	90	97.3%
GPT-5 Nano	100	100	99	96	91	97.2%
Claude Sonnet 4.6	100	100	98	97	90	97.1%
Xiaomi MIMO v2.5 Pro	100	100	97	95	93	97.1%
Mistral Small 4	100	100	98	97	91	97.1%
Stealth: Hunter Alpha	98	98	98	96	95	97.0%
Ministral 3B	100	100	99	96	90	97.0%
Claude Haiku 4.5	100	100	100	98	87	97.0%
Mistral Small Creative	100	98	97	96	92	96.8%
Z.AI GLM 4.7 Flash	100	100	96	94	94	96.8%
Claude 3.7 Sonnet	100	100	98	97	89	96.8%
Hermes 3 70B	100	100	100	99	83	96.4%
Gemma 4 31B	100	99	98	94	91	96.2%
ByteDance Seed 1.6	100	98	95	95	93	96.2%
GPT-4o, May 13th (temp=0)	100	97	96	94	93	96.0%
Aion 2.0	100	100	99	92	89	95.9%
Arcee AI: Trinity Mini	100	100	100	92	88	95.9%
Z.AI GLM 5	100	100	95	94	90	95.7%
DeepSeek V4 Pro (Reasoning)	100	100	100	98	80	95.6%
Claude Opus 4.7	100	100	94	93	90	95.3%
Llama 3.1 Nemotron 70B	100	100	96	96	85	95.2%
DeepSeek V3.1	100	97	95	94	89	95.2%
ByteDance Seed 2.0 Lite	100	99	96	96	85	95.0%
Gemini 3.1 Flash Lite	96	96	96	95	92	95.0%
Stealth: Healer Alpha	100	100	96	92	87	95.0%
Gemini 3.1 Flash Lite (Reasoning)	100	97	96	95	87	94.9%
Mistral Large 3	99	98	95	94	89	94.9%
Gemini 2.5 Pro	100	100	97	91	87	94.8%
Gemini 2.5 Flash Lite (Reasoning)	100	100	99	89	85	94.6%
Ministral 3 8B	99	97	96	94	86	94.4%
Qwen 2.5 72B	98	96	94	92	91	94.3%
Gemma 3 27B	100	100	93	92	85	94.1%
Ministral 8B	100	97	95	92	86	94.0%
Mistral Large 2	100	93	93	92	90	93.7%
Nemotron 3 Nano	100	100	100	97	70	93.4%
Llama 3.1 8B	100	98	93	88	87	93.3%
Gemini 3 Flash (Preview)	99	99	94	88	86	93.2%
Z.AI GLM 4.6	100	99	94	86	86	93.1%
Inception Mercury	100	100	100	89	75	92.8%
GPT-4o, Aug. 6th (temp=0)	100	100	100	89	73	92.5%
Gemini 2.5 Flash Lite	94	93	92	89	87	91.0%
Mistral NeMO	98	93	85	85	78	87.9%
ByteDance Seed 2.0 Mini	94	89	69	67	66	77.0%

genre

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
GPT-5.5 (Reasoning)	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%
Claude Opus 4	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Claude Haiku 4.5	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%
Claude Opus 4.7 (Reasoning)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	99.9%
GPT-4.1 Mini	100	100	100	100	100	99.9%
Mistral Small 4	100	100	100	100	100	99.9%
Mistral Small 4 (Reasoning)	100	100	100	100	99	99.9%
GPT-4.1 Nano	100	100	100	100	99	99.9%
Writer: Palmyra X5	100	100	100	100	99	99.8%
Grok 4.20	100	100	100	100	99	99.8%
Qwen 3.5 Plus (2026-04-20)	100	100	100	99	99	99.7%
GPT-4o, Aug. 6th (temp=0)	100	100	100	99	99	99.6%
Grok 4.3 (Reasoning)	100	100	100	100	98	99.6%
Grok 4	100	100	100	100	98	99.6%
Gemma 4 31B	100	100	99	99	99	99.5%
Qwen 3.5 122B	100	100	100	100	98	99.5%
o4 Mini High	100	100	100	100	98	99.5%
Claude 3.7 Sonnet	100	100	100	99	99	99.5%
Qwen3.6 Max Preview	100	100	100	100	97	99.3%
GPT-5.4	100	100	100	100	96	99.3%
GPT-4o, Aug. 6th (temp=1)	100	100	100	99	98	99.3%
Grok 4.3	100	100	100	98	98	99.2%
GPT-5.5 (Reasoning, Low)	100	100	100	100	96	99.2%
Qwen 3.5 27B	100	100	100	98	98	99.1%
Mistral Large 3	100	100	100	100	95	99.1%
GPT-5.4 (Reasoning, Low)	100	100	100	98	97	99.0%
GPT-4o Mini (temp=0)	100	100	100	100	96	99.0%
MiniMax M2.5	100	100	100	100	95	99.0%
Z.AI GLM 4.7	100	100	100	99	96	99.0%
GPT-5.5	100	100	100	98	97	99.0%
GPT-5.4 (Reasoning)	100	100	99	98	97	98.9%
Grok 4.20 (Reasoning)	100	100	99	98	98	98.9%
Qwen 3.5 Plus (2026-02-15)	100	100	100	97	97	98.8%
Z.AI GLM 5	100	100	100	97	96	98.7%
GPT-5.1	100	100	99	97	97	98.6%
Mistral Large 2	100	100	100	97	96	98.6%
Ministral 3B	100	100	98	97	97	98.6%
Qwen 3 32B	100	100	100	98	95	98.6%
MoonshotAI: Kimi K2.6	100	100	100	97	96	98.6%
Grok 4.20 (Beta, Reasoning)	100	100	100	97	96	98.6%
Gemini 2.5 Flash Lite (Reasoning)	100	100	99	99	95	98.6%
Gemma 3 27B	100	100	100	100	93	98.5%
Mistral Small Creative	100	100	98	97	97	98.5%
Gemini 3 Pro (Preview)	100	100	100	100	92	98.4%
Claude Opus 4.6	100	99	99	99	94	98.3%
ByteDance Seed 1.6 Flash	100	100	100	99	93	98.3%
Claude Sonnet 4	100	100	100	99	93	98.3%
GPT-OSS 120B	100	99	99	98	96	98.3%
Qwen 3.6 Flash	100	100	100	97	94	98.1%
Claude Opus 4.5	100	100	100	97	93	98.1%
Qwen 3.5 397B A17B	100	100	100	100	90	98.1%
GPT-4.1	100	100	100	100	90	98.1%
Z.AI GLM 4.5 Air	100	100	100	100	90	98.1%
Claude 3 Haiku	100	100	99	99	93	98.1%
Nemotron 3 Nano	100	100	98	98	94	98.1%
Mistral Large	100	100	100	99	91	98.1%
DeepSeek V4 Flash (Reasoning)	100	100	100	97	93	98.0%
Claude Sonnet 4.5	100	100	100	100	90	98.0%
GPT-5.2	100	99	99	96	96	98.0%
Z.AI GLM 5.1	100	100	100	97	93	97.9%
GPT-5	100	100	100	97	93	97.9%
MoonshotAI: Kimi K2.5	100	100	100	95	94	97.9%
GPT-4o, May 13th (temp=1)	100	100	100	97	92	97.8%
Arcee AI: Trinity Mini	100	100	100	100	89	97.7%
DeepSeek-V2 Chat	100	100	100	95	94	97.7%
Mistral NeMO	100	100	100	98	90	97.7%
ByteDance Seed 1.6	100	100	100	96	91	97.5%
Gemini 2.5 Flash	100	100	99	96	92	97.3%
Claude Opus 4.6 (Reasoning)	100	100	100	98	89	97.3%
Qwen 3.5 9B	100	100	100	96	90	97.2%
Claude 3.5 Sonnet	100	100	100	98	87	97.0%
GPT-5.4 Nano (Reasoning, Low)	100	99	99	95	91	97.0%
DeepSeek V4 Pro	100	100	98	94	93	96.9%
Gemma 3 4B	99	99	99	94	93	96.9%
Rocinante 12B	100	100	97	94	94	96.8%
Nemotron 3 Super	100	100	98	93	93	96.8%
Gemini 2.5 Flash Lite	100	100	98	94	92	96.8%
DeepSeek V4 Flash	100	100	96	96	92	96.7%
Hermes 3 70B	100	100	98	93	93	96.6%
Mistral Medium 3.1	100	100	100	94	89	96.6%
Cohere Command R+ (Aug. 2024)	100	100	99	93	92	96.6%
GPT-5.4 Mini	100	100	98	93	91	96.5%
LFM2 24B	100	99	98	94	92	96.5%
Qwen 3.6 35B	100	100	100	94	89	96.5%
Z.AI GLM 5 Turbo	100	100	100	97	85	96.4%
GPT-5 Nano	100	99	97	95	91	96.4%
Grok 4.20 (Beta)	100	100	99	93	90	96.4%
Gemini 3.1 Pro (Preview)	100	100	97	95	89	96.3%
Z.AI GLM 4.7 Flash	100

Passive voice 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