POV shift: 3rd person to 1st person (Elena's perspective)

Text Replacement

Tests deterministic text transformations: renaming characters/locations, expanding contractions, tense rewriting, POV shifts, gender swaps, combined transformations, and word avoidance. Scored by checking each expected change independently.

Text Editing

Performance Score Distribution (Top 20)

Click a model name to view its detail page.

	Score
GPT-5.6 Sol (Reasoning)	100%
Claude Opus 4.6 (Reasoning)	100%
Qwen3.7 Max	100%
Grok 4.5 (Reasoning, High)	100%
Gemini 3.1 Pro (Preview)	100%
GPT-5.4 (Reasoning)	100%
Z.AI GLM 5.1	100%
Qwen3.6 Max Preview	100%
GPT-5.5 (Reasoning)	100%
Claude Sonnet 4.6 (Reasoning)	100%
Z.AI GLM 5.2 (Reasoning, High)	100%
Gemini 3.5 Flash (Reasoning)	100%
Z.AI GLM 5 Turbo	100%
MoonshotAI: Kimi K2.6	100%
Claude Opus 4.7 (Reasoning)	100%
GPT-5.5 (Reasoning, Low)	100%
GPT-5.6 Terra (Reasoning)	100%
Claude Opus 4.8 (Reasoning)	100%
Claude Opus 4.6	100%
Claude Opus 4.8 (Reasoning, Low)	100%

	Score	Cost	Time
Ministral 3B	99%	$0.0000	2.2s
Mistral NeMO	100%	$0.0002	2.0s
Gemini 2.5 Flash Lite	100%	$0.0003	1.6s
Ministral 3 3B	100%	$0.0001	2.4s
Ministral 8B	100%	$0.0001	3.5s
Ministral 3 8B	100%	$0.0002	3.0s
Mistral Small 4	100%	$0.0004	3.3s
Ministral 3 14B	100%	$0.0002	3.8s
GPT-4.1 Nano	100%	$0.0003	3.5s
Gemini 3.1 Flash Lite (Preview)	100%	$0.0009	1.8s
Gemini 3.1 Flash Lite (Reasoning)	100%	$0.0009	1.7s
Gemini 3.1 Flash Lite	100%	$0.0009	3.1s
Mistral Small 3.2 24B	100%	$0.0002	4.8s
Gemma 3 4B	100%	$0.0001	6.4s
GPT-5.4 Nano (Reasoning)	100%	$0.0009	3.5s
Gemini 2.5 Flash	100%	$0.0015	2.1s
DeepSeek V4 Flash	100%	$0.0002	6.5s
Inception Mercury 2	100%	$0.0018	2.3s
Mistral Medium 3.1	100%	$0.0013	6.2s
Gemma 4 26B	100%	$0.0003	11.4s

	Score	Cost	Speed	Stability
Gemini 2.5 Flash Lite	100%	$0.0003	1.6s	100%
Mistral NeMO	100%	$0.0002	2.0s	100%
Ministral 3 8B	100%	$0.0002	3.0s	100%
Ministral 8B	100%	$0.0001	3.5s	100%
Gemini 3.1 Flash Lite (Reasoning)	100%	$0.0009	1.7s	100%
Gemini 3.1 Flash Lite (Preview)	100%	$0.0009	1.8s	100%
Mistral Small 4	100%	$0.0004	3.3s	100%
Ministral 3 14B	100%	$0.0002	3.8s	100%
Mistral Small 3.2 24B	100%	$0.0002	4.8s	100%
Gemini 3.1 Flash Lite	100%	$0.0009	3.1s	100%
Gemini 2.5 Flash	100%	$0.0015	2.1s	100%
Ministral 3 3B	100%	$0.0001	2.4s	99%
Inception Mercury 2	100%	$0.0018	2.3s	100%
Gemma 3 4B	100%	$0.0001	6.4s	100%
DeepSeek V4 Flash	100%	$0.0002	6.5s	100%
Gemini 3 Flash (Preview)	100%	$0.0018	3.1s	100%
GPT-5.4 Mini	100%	$0.0027	1.9s	100%
GPT-4.1 Nano	100%	$0.0003	3.5s	98%
GPT-4o Mini (temp=0)	100%	$0.0004	8.0s	100%
GPT-4o Mini (temp=1)	100%	$0.0004	8.1s	100%

Median	Evaluator	Top 3	Flop 3
100.0%	Dialogue and other-character pronouns preserved	100GPT-5.4 Nano (Reasoning, Low) 100Gemma 3 4B 100Claude Opus 4.5	29Hermes 3 70B 73DeepSeek V3.1 75DeepSeek V3 (2025-03-24)
100.0%	No remaining protagonist name in narration	100GPT-5.1 100Grok 4.3 100Qwen 3.6 35B	71Gemini 3.5 Flash (Reasoning, Minimal) 71Hermes 3 70B 86DeepSeek V3 (2025-03-24)
100.0%	POV transformation accuracy	100Claude Opus 4.7 (Reasoning) 100GPT-4.1 100Aion 3.0	29Hermes 3 70B 74DeepSeek V3 (2025-03-24) 74DeepSeek V3.1

Text Replacement

POV shift: 3rd person to 1st person (Elena's perspective)

Performance Score Distribution (Top 20)

Price-Performance Score Distribution (Top 20)

Most Stable Models (Top 20)

Top Overall Models (Top 20)

Rank	Model	Avg. Cost	Avg. Time	Stability	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Total
91	GPT-5.6 Sol (Reasoning)	$0.019	3.8s	100%	100	100	100	100	100	100	100	100%
129	Claude Opus 4.6 (Reasoning)	$0.043	19.4s	100%	100	100	100	100	100	100	100	100%
128	Qwen3.7 Max	$0.029	49.9s	100%	100	100	100	100	100	100	100	100%
107	Grok 4.5 (Reasoning, High)	$0.016	25.3s	100%	100	100	100	100	100	100	100	100%
143	Gemini 3.1 Pro (Preview)	$0.075	1.2m	100%	100	100	100	100	100	100	100	100%
102	GPT-5.4 (Reasoning)	$0.018	17.0s	100%	100	100	100	100	100	100	100	100%
136	Z.AI GLM 5.1	$0.015	1.9m	100%	100	100	100	100	100	100	100	100%
139	Qwen3.6 Max Preview	$0.034	1.8m	100%	100	100	100	100	100	100	100	100%
111	GPT-5.5 (Reasoning)	$0.026	9.3s	100%	100	100	100	100	100	100	100	100%
135	Claude Sonnet 4.6 (Reasoning)	$0.045	28.5s	100%	100	100	100	100	100	100	100	100%
74	Z.AI GLM 5.2 (Reasoning, High)	$0.0073	20.2s	100%	100	100	100	100	100	100	100	100%
113	Gemini 3.5 Flash (Reasoning)	$0.028	11.7s	100%	100	100	100	100	100	100	100	100%
77	Z.AI GLM 5 Turbo	$0.0089	17.4s	100%	100	100	100	100	100	100	100	100%
141	MoonshotAI: Kimi K2.6	$0.022	2.3m	100%	100	100	100	100	100	100	100	100%
99	Claude Opus 4.7 (Reasoning)	$0.023	3.8s	100%	100	100	100	100	100	100	100	100%
97	GPT-5.5 (Reasoning, Low)	$0.021	7.4s	100%	100	100	100	100	100	100	100	100%
58	GPT-5.6 Terra (Reasoning)	$0.0095	3.3s	100%	100	100	100	100	100	100	100	100%
108	Claude Opus 4.8 (Reasoning)	$0.024	8.1s	100%	100	100	100	100	100	100	100	100%
87	Claude Opus 4.6	$0.017	5.6s	100%	100	100	100	100	100	100	100	100%
106	Claude Opus 4.8 (Reasoning, Low)	$0.024	7.6s	100%	100	100	100	100	100	100	100	100%
127	GPT-5	$0.031	44.8s	100%	100	100	100	100	100	100	100	100%
79	GPT-5 Mini	$0.0049	27.9s	100%	100	100	100	100	100	100	100	100%
133	Qwen 3.5 397B A17B	$0.0099	1.9m	100%	100	100	100	100	100	100	100	100%
88	Grok 4.3 (Reasoning)	$0.0068	31.3s	100%	100	100	100	100	100	100	100	100%
80	Grok 4.5 (Reasoning, Low)	$0.011	13.8s	100%	100	100	100	100	100	100	100	100%
61	GPT-5.4 (Reasoning, Low)	$0.0094	5.3s	100%	100	100	100	100	100	100	100	100%
72	Grok 4.20 (Reasoning)	$0.0056	18.4s	100%	100	100	100	100	100	100	100	100%
138	MoonshotAI: Kimi K2.5	$0.014	2.4m	100%	100	100	100	100	100	100	100	100%
92	GPT-5.6 Sol	$0.018	4.7s	100%	100	100	100	100	100	100	100	100%
84	GPT-5.1	$0.013	13.3s	100%	100	100	100	100	100	100	100	100%
64	Claude Sonnet 4.6	$0.010	4.2s	100%	100	100	100	100	100	100	100	100%
35	GPT-5.6 Luna (Reasoning)	$0.0041	2.9s	100%	100	100	100	100	100	100	100	100%
81	MiniMax M3	$0.0023	35.8s	100%	100	100	100	100	100	100	100	100%
70	Claude Sonnet 5 (Reasoning)	$0.0096	8.4s	100%	100	100	100	100	100	100	100	100%
145	Qwen 3.5 122B	$0.055	2.7m	100%	100	100	100	100	100	100	100	100%
71	Claude Sonnet 5 (Reasoning, Low)	$0.0097	8.4s	100%	100	100	100	100	100	100	100	100%
122	Qwen 3.5 27B	$0.016	1.1m	100%	100	100	100	100	100	100	100	100%
98	Gemini 3 Flash (Preview, Reasoning)	$0.010	34.0s	100%	100	100	100	100	100	100	100	100%
101	Claude Opus 4.7	$0.023	4.6s	100%	100	100	100	100	100	100	100	100%
46	GPT-5.4 Mini (Reasoning)	$0.0055	4.3s	100%	100	100	100	100	100	100	100	100%
120	Qwen 3.5 Plus (2026-04-20)	$0.011	1.2m	100%	100	100	100	100	100	100	100	100%
126	Gemma 4 31B (Reasoning)	$0.0011	1.8m	100%	100	100	100	100	100	100	100	100%
86	Claude Opus 4.5	$0.017	4.8s	100%	100	100	100	100	100	100	100	100%
100	Z.AI GLM 5	$0.0086	39.9s	100%	100	100	100	100	100	100	100	100%
93	ByteDance Seed 1.6	$0.0038	40.8s	100%	100	100	100	100	100	100	100	100%
55	GPT-5.2	$0.0081	5.3s	100%	100	100	100	100	100	100	100	100%
90	GPT-5.5	$0.018	4.3s	100%	100	100	100	100	100	100	100	100%
78	Qwen 3.6 Flash	$0.0074	21.5s	100%	100	100	100	100	100	100	100	100%
134	DeepSeek V4 Pro (Reasoning)	$0.0061	2.0m	100%	100	100	100	100	100	100	100	100%
123	Gemma 4 26B (Reasoning)	$0.0013	1.8m	100%	100	100	100	100	100	100	100	100%
125	Aion 3.0	$0.021	1.0m	100%	100	100	100	100	100	100	100	100%
116	o4 Mini High	$0.021	35.9s	100%	100	100	100	100	100	100	100	100%
112	Gemini 2.5 Pro	$0.024	17.4s	100%	100	100	100	100	100	100	100	100%
119	Qwen 3.6 27B	$0.014	59.9s	100%	100	100	100	100	100	100	100	100%
56	GPT-5.6 Terra	$0.0091	3.1s	100%	100	100	100	100	100	100	100	100%
53	DeepSeek V4 Flash (Reasoning)	$0.0005	20.4s	100%	100	100	100	100	100	100	100	100%
118	Z.AI GLM 4.7	$0.0052	1.3m	100%	100	100	100	100	100	100	100	100%
76	Qwen 3.6 35B	$0.0047	27.2s	100%	100	100	100	100	100	100	100	100%
85	Z.AI GLM 4.6	$0.0045	35.3s	100%	100	100	100	100	100	100	100	100%
68	Claude Sonnet 4	$0.010	5.9s	100%	100	100	100	100	100	100	100	100%
65	Claude Sonnet 4.5	$0.010	4.7s	100%	100	100	100	100	100	100	100	100%
69	Claude Sonnet 5	$0.0094	8.7s	100%	100	100	100	100	100	100	100	100%
132	Claude Opus 4	$0.052	7.4s	100%	100	100	100	100	100	100	100	100%
115	Qwen 3.5 35B	$0.015	47.0s	100%	100	100	100	100	100	100	100	100%
43	GPT-4.1	$0.0053	3.9s	100%	100	100	100	100	100	100	100	100%
105	MiniMax M2.5	$0.0017	59.6s	100%	100	100	100	100	100	100	100	100%
94	Aion 2.0	$0.0038	44.7s	100%	100	100	100	100	100	100	100	100%
95	o4 Mini	$0.014	20.7s	100%	100	100	100	100	100	100	100	100%
131	MiniMax M2.7	$0.0087	1.8m	100%	100	100	100	100	100	100	100	100%
29	Qwen 3.5 Plus (2026-02-15)	$0.0015	6.8s	100%	100	100	100	100	100	100	100	100%
63	Xiaomi MIMO v2.5 Pro	$0.0044	18.0s	100%	100	100	100	100	100	100	100	100%
5	Gemini 3.1 Flash Lite (Reasoning)	$0.0009	1.7s	100%	100	100	100	100	100	100	100	100%
130	ByteDance Seed 2.0 Mini	$0.0020	2.0m	100%	100	100	100	100	100	100	100	100%
89	Qwen 3.5 Flash	$0.0026	41.7s	100%	100	100	100	100	100	100	100	100%
16	Gemini 3 Flash (Preview)	$0.0018	3.1s	100%	100	100	100	100	100	100	100	100%
6	Gemini 3.1 Flash Lite (Preview)	$0.0009	1.8s	100%	100	100	100	100	100	100	100	100%
75	Gemma 4 31B	$0.0003	37.7s	100%	100	100	100	100	100	100	100	100%
10	Gemini 3.1 Flash Lite	$0.0009	3.1s	100%	100	100	100	100	100	100	100	100%
33	GPT-5.6 Luna	$0.0036	2.8s	100%	100	100	100	100	100	100	100	100%
66	Z.AI GLM 4.5	$0.0027	23.4s	100%	100	100	100	100	100	100	100	100%
34	Gemma 4 26B	$0.0003	11.4s	100%	100	100	100	100	100	100	100	100%
104	GPT-OSS 120B	$0.0009	1.0m	100%	100	100	100	100	100	100	100	100%
59	GPT-5.4	$0.0091	5.0s	100%	100	100	100	100	100	100	100	100%
28	Mistral Large 3	$0.0011	7.8s	100%	100	100	100	100	100	100	100	100%
103	ByteDance Seed 2.0 Lite	$0.0045	51.0s	100%	100	100	100	100	100	100	100	100%
51	Gemini 2.5 Flash (Reasoning)	$0.0050	7.3s	100%	100	100	100	100	100	100	100	100%
48	DeepSeek-V2 Chat	$0.0008	16.3s	100%	100	100	100	100	100	100	100	100%
121	Qwen 3.5 9B	$0.0013	1.7m	100%	100	100	100	100	100	100	100	100%
62	Xiaomi MIMO v2.5	$0.0045	17.5s	100%	100	100	100	100	100	100	100	100%
110	Aion 3.0 Mini	$0.0037	1.0m	100%	100	100	100	100	100	100	100	100%
26	GPT-5.4 Mini (Reasoning, Low)	$0.0029	2.6s	100%	100	100	100	100	100	100	100	100%
32	Claude Haiku 4.5	$0.0035	2.8s	100%	100	100	100	100	100	100	100	100%
42	Gemini 2.5 Flash Lite (Reasoning)	$0.0015	13.1s	100%	100	100	100	100	100	100	100	100%
44	DeepSeek V3 (2024-12-26)	$0.0007	15.0s	100%	100	100	100	100	100	100	100	100%
82	DeepSeek V3.2	$0.0004	42.3s	100%	100	100	100	100	100	100	100	100%
49	GPT-4o, Aug. 6th (temp=0)	$0.0066	2.3s	100%	100	100	100	100	100	100	100	100%
38	DeepSeek V4 Pro	$0.0005	13.2s	100%	100	100	100	100	100	100	100	100%
15	DeepSeek V4 Flash	$0.0002	6.5s	100%	100	100	100	100	100	100	100	100%
13	Inception Mercury 2	$0.0018	2.3s	100%	100	100	100	100	100	100	100	100%
117	Nemotron 3 Super	$0.0000	1.5m	100%	100	100	100	100	100	100	100	100%
47	Mistral Large 2	$0.0043	7.1s	100%	100	100	100	100	100	100	100	100%
23	GPT-4.1 Mini	$0.0011	6.6s	100%	100	100	100	100	100	100	100	100%
50	GPT-4o, Aug. 6th (temp=1)	$0.0066	3.1s	100%	100	100	100	100	100	100	100	100%
22	Grok 4.20	$0.0020	4.3s	100%	100	100	100	100	100	100	100	100%
60	Hermes 3 405B	$0.0012	24.3s	100%	100	100	100	100	100	100	100	100%
11	Gemini 2.5 Flash	$0.0015	2.1s	100%	100	100	100	100	100	100	100	100%
17	GPT-5.4 Mini	$0.0027	1.9s	100%	100	100	100	100	100	100	100	100%
1	Gemini 2.5 Flash Lite	$0.0003	1.6s	100%	100	100	100	100	100	100	100	100%
45	Qwen3 235B A22B Instruct 2507	$0.0004	16.4s	100%	100	100	100	100	100	100	100	100%
21	Grok 4.3	$0.0020	4.3s	100%	100	100	100	100	100	100	100	100%
20	GPT-4o Mini (temp=1)	$0.0004	8.1s	100%	100	100	100	100	100	100	100	100%
41	Llama 3.1 70B	$0.0005	15.5s	100%	100	100	100	100	100	100	100	100%
9	Mistral Small 3.2 24B	$0.0002	4.8s	100%	100	100	100	100	100	100	100	100%
19	GPT-4o Mini (temp=0)	$0.0004	8.0s	100%	100	100	100	100	100	100	100	100%
25	Mistral Medium 3.1	$0.0013	6.2s	100%	100	100	100	100	100	100	100	100%
27	Gemma 3 12B	$0.0001	10.1s	100%	100	100	100	100	100	100	100	100%
39	Gemma 3 27B	$0.0002	14.9s	100%	100	100	100	100	100	100	100	100%
7	Mistral Small 4	$0.0004	3.3s	100%	100	100	100	100	100	100	100	100%
30	Qwen 2.5 72B	$0.0003	10.2s	100%	100	100	100	100	100	100	100	100%
83	WizardLM 2 8x22b	$0.0008	42.9s	100%	100	100	100	100	100	100	100	100%
8	Ministral 3 14B	$0.0002	3.8s	100%	100	100	100	100	100	100	100	100%
3	Ministral 3 8B	$0.0002	3.0s	100%	100	100	100	100	100	100	100	100%
14	Gemma 3 4B	$0.0001	6.4s	100%	100	100	100	100	100	100	100	100%
2	Mistral NeMO	$0.0002	2.0s	100%	100	100	100	100	100	100	100	100%
4	Ministral 8B	$0.0001	3.5s	100%	100	100	100	100	100	100	100	100%
57	Z.AI GLM 4.5 Air	$0.0010	19.8s	99%	100	100	100	100	100	100	98	100%
67	Qwen 3 32B	$0.0009	24.8s	99%	100	100	100	100	100	100	98	100%
54	Writer: Palmyra X5	$0.0035	11.3s	99%	100	100	100	100	100	100	98	100%
12	Ministral 3 3B	$0.0001	2.4s	99%	100	100	100	100	100	100	98	100%
40	Cydonia 24B V4.1	$0.0004	10.8s	98%	100	100	100	100	100	100	97	100%
18	GPT-4.1 Nano	$0.0003	3.5s	98%	100	100	100	100	100	100	97	100%
24	GPT-5.4 Nano (Reasoning)	$0.0009	3.5s	99%	100	100	100	100	100	99	99	100%
109	GPT-5 Nano	$0.0026	55.6s	98%	100	100	100	100	100	98	98	99%
52	ByteDance Seed 1.6 Flash	$0.0007	12.4s	97%	100	100	100	100	100	99	96	99%
36	GPT-5.4 Nano (Reasoning, Low)	$0.0008	3.7s	97%	100	100	100	99	99	99	99	99%
96	Cohere Command R+ (Aug. 2024)	$0.0070	26.2s	95%	100	100	100	100	100	100	93	99%
37	GPT-5.4 Nano	$0.0008	3.4s	97%	100	100	99	99	99	99	98	99%
31	Ministral 3B	$0.0000	2.2s	97%	100	100	100	100	98	98	96	99%
137	Z.AI GLM 4.7 Flash	$0.0018	1.7m	81%	100	100	100	100	100	100	73	96%
73	Arcee AI: Trinity Mini	$0.0002	6.3s	93%	97	97	97	97	97	95	91	96%
114	Mistral Small 4 (Reasoning)	$0.0015	13.6s	76%	100	100	100	100	100	100	65	95%
142	Nemotron 3 Nano	$0.0020	1.9m	75%	100	100	100	100	99	75	70	92%
124	Gemini 3.5 Flash (Reasoning, Minimal)	$0.0055	2.5s	67%	100	100	100	100	100	65	63	90%
140	DeepSeek V3.1	$0.0005	24.2s	44%	100	100	100	100	100	41	37	82%
144	DeepSeek V3 (2025-03-24)	$0.0007	57.9s	26%	100	100	100	100	100	46	0	78%
146	Hermes 3 70B	$0.0024	3.0m	8%	100	100	33	33	33	2	0	43%
99.08%