Romance: separated couple reunites

Bad Writing Habits

Detects common prose quality anti-patterns in AI-generated creative writing, including passive voice, past progressive overuse, weak dialogue tags, filter words, purple prose, cliches, AI-ism words/adverbs/names, and more.

Creative Writing Hallucination

Performance Score Distribution (Top 20)

Click a model name to view its detail page.

	Score
GPT-5.4 (Reasoning, Low)	90%
GPT-5.4 Mini	90%
GPT-5.4 (Reasoning)	89%
GPT-5.4	89%
Grok 4.20 (Beta, Reasoning)	87%
GPT-5.4 Mini (Reasoning)	87%
GPT-5.1	87%
Mistral Small Creative	86%
GPT-5.4 Mini (Reasoning, Low)	86%
Claude Opus 4	86%
Qwen 3.5 397B A17B	85%
GPT-5	85%
Mistral Small 4 (Reasoning)	85%
Claude Opus 4.6 (Reasoning)	85%
Mistral Large	85%
Mistral Large 3	84%
Qwen 3.5 35B	84%
Z.AI GLM 5	84%
Qwen 3 32B	84%
Qwen 3.5 Flash	84%

	Score	Cost	Time
GPT-5.4 Mini	90%	$0.014	15.5s
Mistral Small Creative	86%	$0.0006	9.0s
Mistral Large 3	84%	$0.0026	28.0s
GPT-5.4 Mini (Reasoning, Low)	86%	$0.015	16.6s
Ministral 3 14B	84%	$0.0004	9.5s
Mistral Small 4	84%	$0.0011	16.9s
GPT-5.4 Mini (Reasoning)	87%	$0.021	24.2s
Qwen 3.5 Flash	84%	$0.0026	45.9s
Qwen 3 32B	84%	$0.0015	1.1m
Qwen 3.5 122B	84%	$0.016	47.3s
ByteDance Seed 1.6 Flash	84%	$0.0012	28.0s
Mistral Small 4 (Reasoning)	85%	$0.0019	29.4s
Mistral Large 2	83%	$0.0097	32.4s
Qwen 3.5 35B	84%	$0.016	1.0m
Mistral Medium 3.1	82%	$0.0040	36.4s
Mistral Large	85%	$0.010	27.4s
Grok 4.20 (Beta, Reasoning)	87%	$0.046	44.0s
Stealth: Healer Alpha	80%	$0.0000	26.3s
LFM2 24B	81%	$0.0001	19.3s
Z.AI GLM 5 Turbo	84%	$0.0085	41.1s

	Score	Consistency	Stability
GPT-5.4 (Reasoning, Low)	90%	99%	90%
GPT-5.4 Mini	90%	98%	88%
GPT-5.4 (Reasoning)	89%	97%	87%
GPT-5.4 Mini (Reasoning)	87%	98%	85%
GPT-5.4 Mini (Reasoning, Low)	86%	98%	84%
GPT-5.4	89%	94%	83%
Mistral Small Creative	86%	96%	83%
Grok 4.20 (Beta, Reasoning)	87%	93%	83%
GPT-5.1	87%	96%	83%
Claude Opus 4	86%	96%	82%
Qwen 3.5 397B A17B	85%	98%	82%
o4 Mini High	84%	98%	82%
Qwen 3.5 Flash	84%	97%	82%
Gemini 3.1 Pro (Preview)	83%	97%	81%
DeepSeek V3 (2025-03-24)	84%	96%	81%
Mistral Large 2	83%	96%	80%
Claude Opus 4.6 (Reasoning)	85%	93%	80%
Qwen 3.5 27B	83%	97%	80%
Mistral Small 4	84%	95%	80%
Qwen 3.5 122B	84%	93%	79%

	Score	Cost	Speed	Stability
GPT-5.4 Mini	90%	$0.014	15.5s	88%
Mistral Small Creative	86%	$0.0006	9.0s	83%
GPT-5.4 Mini (Reasoning, Low)	86%	$0.015	16.6s	84%
GPT-5.4 Mini (Reasoning)	87%	$0.021	24.2s	85%
GPT-5.4 (Reasoning, Low)	90%	$0.052	1.4m	90%
Mistral Small 4	84%	$0.0011	16.9s	80%
Ministral 3 14B	84%	$0.0004	9.5s	79%
Qwen 3.5 Flash	84%	$0.0026	45.9s	82%
ByteDance Seed 1.6 Flash	84%	$0.0012	28.0s	79%
Grok 4.20 (Beta, Reasoning)	87%	$0.046	44.0s	83%
o4 Mini High	84%	$0.021	39.0s	82%
Mistral Large 2	83%	$0.0097	32.4s	80%
Mistral Small 4 (Reasoning)	85%	$0.0019	29.4s	76%
DeepSeek V3 (2025-03-24)	84%	$0.0013	1.1m	81%
Mistral Large	85%	$0.010	27.4s	77%
o4 Mini	83%	$0.012	22.7s	79%
GPT-5.4	89%	$0.051	1.4m	83%
Mistral Large 3	84%	$0.0026	28.0s	75%
Mistral Medium 3.1	82%	$0.0040	36.4s	79%
Qwen 3.5 122B	84%	$0.016	47.3s	79%

Rank	Model	Avg. Cost	Avg. Time	Stability	# 1	# 2	# 3	# 4	# 5	Total
5	GPT-5.4 (Reasoning, Low)	$0.052	1.4m	90%	91	90	90	90	89	90%
1	GPT-5.4 Mini	$0.014	15.5s	88%	91	90	90	89	88	90%
41	GPT-5.4 (Reasoning)	$0.080	2.4m	87%	91	90	89	88	86	89%
17	GPT-5.4	$0.051	1.4m	83%	93	91	89	87	83	89%
10	Grok 4.20 (Beta, Reasoning)	$0.046	44.0s	83%	90	90	89	87	81	87%
4	GPT-5.4 Mini (Reasoning)	$0.021	24.2s	85%	88	88	87	87	85	87%
29	GPT-5.1	$0.047	1.5m	83%	90	87	87	85	83	87%
2	Mistral Small Creative	$0.0006	9.0s	83%	89	88	87	85	83	86%
3	GPT-5.4 Mini (Reasoning, Low)	$0.015	16.6s	84%	88	87	86	86	84	86%
102	Claude Opus 4	$0.175	1.2m	82%	89	87	86	86	82	86%
90	Qwen 3.5 397B A17B	$0.0068	5.4m	82%	87	85	84	84	84	85%
67	GPT-5	$0.058	1.8m	78%	89	87	83	82	81	85%
13	Mistral Small 4 (Reasoning)	$0.0019	29.4s	76%	90	88	83	83	79	85%
64	Claude Opus 4.6 (Reasoning)	$0.076	1.3m	80%	88	87	86	82	80	85%
15	Mistral Large	$0.010	27.4s	77%	91	87	83	81	81	85%
18	Mistral Large 3	$0.0026	28.0s	75%	91	91	87	78	75	84%
24	Qwen 3.5 35B	$0.016	1.0m	79%	88	87	84	83	80	84%
23	Z.AI GLM 5	$0.0073	59.0s	78%	90	83	83	83	82	84%
45	Qwen 3 32B	$0.0015	1.1m	71%	92	92	83	81	73	84%
8	Qwen 3.5 Flash	$0.0026	45.9s	82%	86	85	84	83	82	84%
6	Mistral Small 4	$0.0011	16.9s	80%	88	84	84	83	82	84%
49	ByteDance Seed 2.0 Lite	$0.012	2.3m	79%	87	86	85	83	77	84%
30	Z.AI GLM 5 Turbo	$0.0085	41.1s	75%	92	85	83	80	79	84%
20	Qwen 3.5 122B	$0.016	47.3s	79%	88	86	86	81	78	84%
11	o4 Mini High	$0.021	39.0s	82%	85	84	84	83	83	84%
9	ByteDance Seed 1.6 Flash	$0.0012	28.0s	79%	88	84	83	82	81	84%
7	Ministral 3 14B	$0.0004	9.5s	79%	88	85	83	81	81	84%
14	DeepSeek V3 (2025-03-24)	$0.0013	1.1m	81%	86	84	84	82	82	84%
43	Claude Sonnet 4.5	$0.029	33.6s	75%	92	83	83	80	79	83%
12	Mistral Large 2	$0.0097	32.4s	80%	85	85	83	83	80	83%
26	Writer: Palmyra X5	$0.011	22.3s	76%	89	83	82	81	79	83%
21	Qwen3 235B A22B Instruct 2507	$0.0010	53.3s	78%	87	86	84	79	79	83%
77	Claude Opus 4.6	$0.071	1.1m	77%	89	86	85	78	76	83%
94	Gemini 3.1 Pro (Preview)	$0.110	1.5m	81%	85	84	83	82	80	83%
35	Qwen 3.5 9B	$0.0010	1.4m	77%	88	85	82	81	79	83%
16	o4 Mini	$0.012	22.7s	79%	87	83	83	82	79	83%
31	Qwen 3.5 27B	$0.011	1.2m	80%	85	83	82	82	81	83%
42	Claude 3.5 Sonnet	$0.036	30.1s	77%	87	85	83	80	77	82%
25	Grok 4.20 (Beta)	$0.018	16.4s	78%	86	84	83	79	79	82%
19	Mistral Medium 3.1	$0.0040	36.4s	79%	84	84	84	82	76	82%
111	MoonshotAI: Kimi K2.5	$0.022	5.7m	79%	85	81	81	81	81	82%
56	Claude Sonnet 4	$0.028	40.0s	74%	89	83	81	79	77	82%
75	Gemini 3 Pro (Preview)	$0.055	57.9s	76%	87	81	81	80	78	81%
52	MiniMax M2.5	$0.0030	1.3m	75%	87	83	81	80	76	81%
22	GPT-5.4 Nano (Reasoning, Low)	$0.0070	24.9s	79%	83	82	81	80	80	81%
44	DeepSeek V3.2	$0.0011	1.3m	77%	84	83	81	81	76	81%
39	Ministral 8B	$0.0002	7.8s	72%	89	84	81	80	73	81%
82	WizardLM 2 8x22b	$0.0020	2.9m	75%	86	82	80	79	79	81%
76	GPT-5.2	$0.051	1.4m	78%	83	82	81	81	77	81%
33	Gemma 3 27B	$0.0005	47.9s	77%	84	82	80	80	78	81%
51	Claude 3.7 Sonnet	$0.035	40.6s	78%	83	82	81	79	78	81%
37	GPT-5.4 Nano	$0.0074	33.9s	76%	85	81	80	79	79	81%
40	MiniMax M2.7	$0.0046	1.2m	79%	83	82	81	79	79	81%
27	LFM2 24B	$0.0001	19.3s	76%	85	82	82	78	76	81%
28	Grok 4.1 Fast	$0.0017	35.6s	78%	83	81	80	80	79	81%
62	Z.AI GLM 4.6	$0.0058	58.1s	73%	87	83	79	78	76	81%
58	DeepSeek V3 (2024-12-26)	$0.0018	50.4s	72%	86	85	80	79	71	80%
34	Gemini 3.1 Flash Lite (Preview)	$0.0029	9.0s	75%	85	82	80	80	74	80%
46	GPT-4.1	$0.016	43.7s	77%	83	82	81	78	76	80%
32	Stealth: Healer Alpha	$0.0000	26.3s	77%	82	82	82	78	75	80%
70	Gemini 2.5 Pro	$0.035	36.8s	74%	85	81	80	79	74	80%
47	Z.AI GLM 4.5	$0.0049	46.5s	76%	84	80	79	78	77	80%
85	Claude Opus 4.5	$0.069	56.6s	78%	81	80	80	79	77	80%
50	Stealth: Hunter Alpha	$0.0000	1.0m	76%	83	81	80	76	76	79%
117	ByteDance Seed 2.0 Mini	$0.0045	5.1m	63%	91	86	75	73	72	79%
57	Z.AI GLM 4.7	$0.0098	56.1s	76%	83	81	80	77	75	79%
38	Grok 4 Fast	$0.0016	22.8s	77%	82	79	79	78	77	79%
36	Ministral 3 8B	$0.0004	8.3s	75%	83	80	80	78	75	79%
59	GPT-5 Mini	$0.010	1.2m	77%	81	79	79	79	77	79%
72	Claude Sonnet 4.6 (Reasoning)	$0.031	42.1s	75%	83	80	79	77	75	79%
63	Aion 2.0	$0.0062	1.4m	77%	80	80	80	79	76	79%
48	Qwen 3.5 Plus (2026-02-15)	$0.0052	32.8s	75%	82	80	79	76	75	79%
65	Claude Haiku 4.5	$0.0098	19.9s	71%	86	80	78	77	72	79%
107	Grok 4	$0.054	2.1m	75%	81	80	78	78	75	78%
71	Rocinante 12B	$0.0015	48.0s	71%	83	82	77	76	74	78%
61	Gemini 3 Flash (Preview, Reasoning)	$0.013	31.8s	74%	82	80	79	76	75	78%
68	GPT-4o, May 13th (temp=1)	$0.024	12.8s	73%	82	80	80	79	70	78%
91	Hermes 3 405B	$0.0021	45.8s	66%	87	85	76	75	68	78%
53	GPT-5.4 Nano (Reasoning)	$0.0077	28.5s	76%	80	79	78	77	76	78%
54	GPT-4o Mini (temp=1)	$0.0011	33.7s	75%	81	80	80	76	73	78%
83	Cohere Command R+ (Aug. 2024)	$0.015	52.9s	72%	84	79	79	73	72	77%
110	ByteDance Seed 1.6	$0.015	2.9m	69%	83	82	76	73	73	77%
55	Ministral 3 3B	$0.0003	6.4s	73%	81	78	76	76	74	77%
60	Ministral 3B	$0.0002	14.6s	72%	82	79	78	74	73	77%
80	Hermes 3 70B	$0.0006	1.2m	72%	82	79	78	75	71	77%
98	DeepSeek-V2 Chat	$0.0017	1.1m	67%	85	79	74	73	71	77%
100	Llama 3.1 8B	$0.0001	24.9s	62%	89	82	74	70	68	77%
96	DeepSeek V3.1	$0.0020	2.1m	74%	78	77	77	75	74	76%
66	Claude 3.5 Haiku	$0.0020	7.4s	72%	80	76	76	76	71	76%
86	Claude Sonnet 4.6	$0.024	32.3s	73%	78	77	76	75	72	76%
92	Arcee AI: Trinity Large (Preview)	$0.0000	52.1s	69%	80	79	74	73	72	76%
81	Qwen 2.5 72B	$0.0006	31.5s	70%	81	76	74	74	73	75%
69	Gemini 2.5 Flash	$0.0047	10.0s	72%	78	77	76	75	71	75%
89	Llama 3.1 Nemotron 70B	$0.0024	29.5s	68%	82	78	74	72	72	75%
73	GPT-4o Mini (temp=0)	$0.0011	33.8s	73%	77	77	76	74	72	75%
79	GPT-4o, May 13th (temp=0)	$0.023	10.7s	73%	77	75	75	74	73	75%
101	GPT-4o, Aug. 6th (temp=1)	$0.016	29.4s	67%	80	80	74	72	69	75%
95	Z.AI GLM 4.7 Flash	$0.0016	59.9s	70%	79	78	73	73	72	75%
78	Mistral NeMO	$0.0003	7.4s	70%	79	77	74	73	72	75%
84	Gemini 3 Flash (Preview)	$0.0083	21.1s	71%	77	76	74	74	71	74%
74	Inception Mercury 2	$0.0025	6.5s	72%	76	76	76	72	70	74%
93	Gemma 3 12B	$0.0003	37.0s	70%	76	75	75	74	67	74%
87	Gemini 2.5 Flash (Reasoning)	$0.012	22.7s	72%	75	74	73	73	72	74%
88	GPT-4.1 Nano	$0.0006	13.6s	69%	76	76	74	73	68	73%
99	GPT-4.1 Mini	$0.0021	15.6s	67%	78	76	73	70	67	73%
97	Stealth: Aurora Alpha	$0.0000	7.4s	67%	78	73	72	72	70	73%
104	Inception Mercury	$0.011	16.0s	67%	77	74	73	70	66	72%
106	Gemini 2.5 Flash Lite (Reasoning)	$0.0030	39.4s	68%	76	74	73	70	66	72%
103	Claude 3 Haiku	$0.0023	11.5s	66%	77	74	71	68	67	71%
114	GPT-5 Nano	$0.0047	1.5m	64%	78	72	69	68	68	71%
112	Llama 3.1 70B	$0.0008	21.5s	60%	80	76	68	67	64	71%
105	Gemma 3 4B	$0.0002	19.7s	67%	74	74	72	67	66	70%
118	Mistral Small 3.2 24B	$0.0044	3.7m	63%	79	71	70	67	64	70%
113	GPT-4o, Aug. 6th (temp=0)	$0.015	15.4s	63%	74	72	71	69	60	69%
109	Gemini 2.5 Flash Lite	$0.0009	10.3s	65%	74	72	72	69	60	69%
115	Nemotron 3 Super	$0.0000	2.1m	64%	75	71	71	67	61	69%
108	Arcee AI: Trinity Mini	$0.0003	9.4s	67%	71	70	69	67	67	69%
116	Nemotron 3 Nano	$0.0009	1.3m	61%	72	70	66	64	64	67%
79.47%

Median	Evaluator	Top 3	Flop 3
100.0%	"Not X but Y" pattern overuse	100Gemma 3 12B 100GPT-5 Mini 100Llama 3.1 8B	21GPT-5 Nano 34Nemotron 3 Super 71Gemma 3 4B
45.1%	Adverb-first sentence starts	97GPT-5.4 Mini 94GPT-5.4 (Reasoning, Low) 93Mistral Medium 3.1	0GPT-4o Mini (temp=0) 0Inception Mercury 2 0Gemini 3.1 Pro (Preview)
100.0%	Adverbs in dialogue tags	100Claude Opus 4.6 (Reasoning) 100Gemini 3.1 Pro (Preview) 100Qwen 3.5 9B	34Llama 3.1 70B 38Llama 3.1 8B 41Hermes 3 70B
88.5%	AI-ism adverb frequency	100Gemini 3.1 Pro (Preview) 99GPT-5.2 98Grok 4.1 Fast	51Claude 3.5 Haiku 55Claude 3 Haiku 58Mistral Small 3.2 24B
100.0%	AI-ism character names	100Mistral NeMO 100Ministral 3 8B 100Inception Mercury	88Claude Sonnet 4 92DeepSeek V3.2 92Claude 3.7 Sonnet
100.0%	AI-ism location names	100Z.AI GLM 4.5 100Cohere Command R+ (Aug. 2024) 100GPT-4.1 Nano	—
47.1%	AI-ism word frequency	89ByteDance Seed 2.0 Mini 85ByteDance Seed 2.0 Lite 84GPT-5.4	0GPT-4o, Aug. 6th (temp=0) 0GPT-4o Mini (temp=0) 0GPT-4o Mini (temp=1)
93.3%	Cliché density	100GPT-5.4 (Reasoning, Low) 100Grok 4.20 (Beta) 100Z.AI GLM 5	33GPT-4o, Aug. 6th (temp=0) 40Gemini 2.5 Flash Lite 47Qwen 2.5 72B
95.5%	Dialogue tag variety (said vs. fancy)	100Mistral Large 2 100GPT-5.4 Nano 100Mistral Medium 3.1	13Hermes 3 70B 24Claude 3 Haiku 25GPT-4o, Aug. 6th (temp=1)
56.0%	Em-dash & semicolon overuse	100Qwen 3.5 122B 100GPT-5.4 Mini 100Z.AI GLM 4.6	0Grok 4 Fast 0Grok 4.1 Fast 0Claude Opus 4.6 (Reasoning)
100.0%	Emotion telling (show vs. tell)	100Qwen 3.5 9B 100Gemini 3 Pro (Preview) 100Grok 4	85GPT-4o, May 13th (temp=0) 89Mistral Small 3.2 24B 90Mistral NeMO
98.3%	Filter word density	100Mistral Small 4 (Reasoning) 100GPT-5 100ByteDance Seed 1.6 Flash	42Llama 3.1 Nemotron 70B 51Nemotron 3 Nano 61Llama 3.1 70B
100.0%	Gibberish response detection	100Qwen 3.5 35B 100Gemini 2.5 Flash (Reasoning) 100Hermes 3 405B	72Llama 3.1 8B 81Llama 3.1 70B 99MiniMax M2.5
100.0%	Markdown formatting overuse	100GPT-4o Mini (temp=1) 100Claude Sonnet 4.6 (Reasoning) 100Inception Mercury	80Ministral 3B 80Qwen 3.5 9B 100ByteDance Seed 1.6 Flash
100.0%	Missing dialogue indicators (quotation marks)	100GPT-4.1 Mini 100GPT-4.1 Nano 100Qwen 3.5 Plus (2026-02-15)	71Qwen 3.5 Flash 80Qwen 3.5 35B 91o4 Mini High
80.0%	Name drop frequency	100MiniMax M2.7 100Claude Sonnet 4.6 (Reasoning) 100GPT-4o Mini (temp=1)	0GPT-5.4 Nano (Reasoning, Low) 7GPT-5.4 Nano (Reasoning) 8Ministral 3B
84.6%	Narrator intent-glossing	100Mistral Large 100Gemini 3.1 Pro (Preview) 100Grok 4.20 (Beta, Reasoning)	2GPT-5 Nano 11Nemotron 3 Nano 13Nemotron 3 Super
100.0%	Overuse of "that" (subordinate clause padding)	100Gemini 3 Pro (Preview) 100Gemini 2.5 Flash Lite 100LFM2 24B	40ByteDance Seed 2.0 Mini 69Claude 3.5 Haiku 76Ministral 3 8B
100.0%	Paragraph length variance	100Ministral 3 14B 100Z.AI GLM 5 100Qwen 3.5 35B	38Inception Mercury 39Arcee AI: Trinity Mini 50Llama 3.1 Nemotron 70B
99.6%	Passive voice overuse	100Claude 3.5 Sonnet 100GPT-4o, Aug. 6th (temp=1) 100LFM2 24B	89ByteDance Seed 2.0 Mini 91ByteDance Seed 1.6 92Claude 3 Haiku
100.0%	Past progressive (was/were + -ing) overuse	100ByteDance Seed 1.6 100Nemotron 3 Nano 100GPT-5.4	49Claude Sonnet 4.6 59Claude Sonnet 4.6 (Reasoning) 69MiniMax M2.5
38.2%	Pronoun-first sentence starts	100Ministral 3B 99Llama 3.1 8B 97Llama 3.1 Nemotron 70B	0GPT-5 0GPT-4o Mini (temp=0) 1Gemini 2.5 Flash Lite (Reasoning)
97.2%	Purple prose (modifier overload)	100ByteDance Seed 1.6 Flash 100ByteDance Seed 2.0 Mini 100Grok 4	81GPT-4.1 Nano 83Gemma 3 4B 84Llama 3.1 8B
100.0%	Repeated phrase echo	100GPT-5.1 100Gemini 3 Pro (Preview) 100Inception Mercury 2	—
100.0%	Sentence length variance	100ByteDance Seed 2.0 Mini 100Gemini 3 Pro (Preview) 100GPT-4o Mini (temp=0)	83Inception Mercury 87Stealth: Aurora Alpha 90Inception Mercury 2
50.5%	Sentence opener variety	94Claude 3.5 Haiku 88Llama 3.1 Nemotron 70B 86GPT-4o, Aug. 6th (temp=1)	27Inception Mercury 31Qwen 3.5 9B 33ByteDance Seed 2.0 Lite
15.8%	Subject-first sentence starts	73Rocinante 12B 66Mistral Small Creative 57Mistral Small 4 (Reasoning)	0GPT-5.4 Nano (Reasoning) 0Gemini 3.1 Pro (Preview) 0Gemini 3 Flash (Preview)
20.0%	Subordinate conjunction sentence starts	73Qwen 3.5 122B 71Claude Sonnet 4 65Mistral Small Creative	0LFM2 24B 0Claude Opus 4.5 0Grok 4
83.3%	Technical jargon density	100Claude 3 Haiku 100LFM2 24B 100GPT-4.1 Nano	8GPT-5 Nano 21Claude Sonnet 4.6 27Nemotron 3 Nano
54.4%	Useless dialogue additions	100Gemini 3.1 Flash Lite (Preview) 100Qwen 3.5 9B 100GPT-5.1	0Inception Mercury 0GPT-4o, Aug. 6th (temp=0) 0Gemma 3 4B

Bad Writing Habits

Romance: separated couple reunites

Performance Score Distribution (Top 20)

Price-Performance Score Distribution (Top 20)

Most Stable Models (Top 20)

Top Overall Models (Top 20)