Expand all contractions

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
Claude Opus 4.6 (Reasoning)	100%
Gemini 3.1 Pro (Preview)	100%
Z.AI GLM 5 Turbo	100%
GPT-5.1	100%
Claude Opus 4.6	100%
GPT-5	100%
Qwen 3.5 397B A17B	100%
Qwen 3.5 122B	100%
Grok 4.20 (Beta, Reasoning)	100%
Z.AI GLM 5	100%
MoonshotAI: Kimi K2.5	100%
Qwen 3.5 27B	100%
Gemini 3 Flash (Preview, Reasoning)	100%
GPT-5.2	100%
Claude Opus 4.5	100%
Z.AI GLM 4.6	100%
Gemini 3 Pro (Preview)	100%
Claude Sonnet 4	100%
Z.AI GLM 4.7	100%
Gemini 2.5 Pro	100%

	Score	Cost	Time
Gemini 2.5 Flash Lite	100%	$0.0002	1.3s
Mistral Small 3.2 24B	100%	$0.0002	3.7s
Gemini 3.1 Flash Lite (Preview)	100%	$0.0007	1.5s
Mistral Small 4	99%	$0.0003	2.4s
Stealth: Healer Alpha	99%	$0.0000	7.8s
Mistral NeMO	98%	$0.0001	2.6s
Grok 4 Fast	100%	$0.0007	4.9s
GPT-4o Mini (temp=1)	100%	$0.0003	7.0s
Gemini 2.5 Flash	100%	$0.0012	1.8s
Ministral 8B	95%	$0.0001	2.4s
Inception Mercury	98%	$0.0003	4.9s
GPT-5.4 Nano (Reasoning, Low)	98%	$0.0006	2.4s
Ministral 3 14B	98%	$0.0002	3.2s
Stealth: Hunter Alpha	100%	$0.0000	15.8s
Llama 3.1 8B	98%	$0.0000	6.1s
GPT-5.4 Nano (Reasoning)	98%	$0.0006	2.5s
Inception Mercury 2	97%	$0.0012	1.6s
GPT-4o Mini (temp=0)	100%	$0.0003	8.4s
ByteDance Seed 1.6 Flash	100%	$0.0004	8.4s
Qwen 2.5 72B	100%	$0.0002	7.7s

	Score	Cost	Speed	Stability
Gemini 2.5 Flash Lite	100%	$0.0002	1.3s	100%
Gemini 3.1 Flash Lite (Preview)	100%	$0.0007	1.5s	100%
Mistral Small 3.2 24B	100%	$0.0002	3.7s	100%
Gemini 2.5 Flash	100%	$0.0012	1.8s	100%
Mistral Small 4	99%	$0.0003	2.4s	99%
Gemini 3 Flash (Preview)	100%	$0.0015	2.7s	100%
Grok 4 Fast	100%	$0.0007	4.9s	99%
GPT-4o Mini (temp=1)	100%	$0.0003	7.0s	100%
Mistral Large 3	100%	$0.0008	5.8s	100%
Mistral NeMO	98%	$0.0001	2.6s	97%
Qwen 3.5 Plus (2026-02-15)	100%	$0.0012	5.5s	100%
GPT-4o Mini (temp=0)	100%	$0.0003	8.4s	100%
Gemini 2.5 Flash Lite (Reasoning)	99%	$0.0008	4.6s	99%
GPT-5.4 Nano (Reasoning)	98%	$0.0006	2.5s	97%
Ministral 3 14B	98%	$0.0002	3.2s	97%
GPT-5.4 Nano (Reasoning, Low)	98%	$0.0006	2.4s	97%
ByteDance Seed 1.6 Flash	100%	$0.0004	8.4s	99%
Claude Haiku 4.5	100%	$0.0027	2.2s	100%
GPT-4.1 Mini	99%	$0.0008	5.1s	99%
Mistral Small Creative	97%	$0.0002	2.3s	97%

Rank	Model	Avg. Cost	Avg. Time	Stability	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Total
81	Claude Opus 4.6 (Reasoning)	$0.014	5.1s	100%	100	100	100	100	100	100	100	100%
105	Gemini 3.1 Pro (Preview)	$0.033	29.1s	100%	100	100	100	100	100	100	100	100%
74	Z.AI GLM 5 Turbo	$0.0082	19.2s	100%	100	100	100	100	100	100	100	100%
66	GPT-5.1	$0.0083	6.9s	100%	100	100	100	100	100	100	100	100%
76	Claude Opus 4.6	$0.013	4.9s	100%	100	100	100	100	100	100	100	100%
98	GPT-5	$0.022	32.7s	100%	100	100	100	100	100	100	100	100%
111	Qwen 3.5 397B A17B	$0.0094	2.2m	100%	100	100	100	100	100	100	100	100%
97	Qwen 3.5 122B	$0.018	40.6s	100%	100	100	100	100	100	100	100	100%
80	Grok 4.20 (Beta, Reasoning)	$0.013	7.2s	100%	100	100	100	100	100	100	100	100%
67	Z.AI GLM 5	$0.0041	25.1s	100%	100	100	100	100	100	100	100	100%
112	MoonshotAI: Kimi K2.5	$0.010	2.2m	100%	100	100	100	100	100	100	100	100%
89	Qwen 3.5 27B	$0.011	43.0s	100%	100	100	100	100	100	100	100	100%
71	Gemini 3 Flash (Preview, Reasoning)	$0.0091	14.9s	100%	100	100	100	100	100	100	100	100%
52	GPT-5.2	$0.0063	4.3s	100%	100	100	100	100	100	100	100	100%
75	Claude Opus 4.5	$0.013	4.5s	100%	100	100	100	100	100	100	100	100%
90	Z.AI GLM 4.6	$0.0052	1.1m	100%	100	100	100	100	100	100	100	100%
102	Gemini 3 Pro (Preview)	$0.029	19.4s	100%	100	100	100	100	100	100	100	100%
63	Claude Sonnet 4	$0.0080	5.0s	100%	100	100	100	100	100	100	100	100%
94	Z.AI GLM 4.7	$0.0055	1.2m	100%	100	100	100	100	100	100	100	100%
100	Gemini 2.5 Pro	$0.027	19.1s	100%	100	100	100	100	100	100	100	100%
93	Grok 4	$0.019	22.6s	100%	100	100	100	100	100	100	100	100%
60	Claude Sonnet 4.5	$0.0080	3.9s	100%	100	100	100	100	100	100	100	100%
106	Claude Opus 4	$0.040	6.5s	100%	100	100	100	100	100	100	100	100%
36	Gemini 2.5 Flash (Reasoning)	$0.0035	5.2s	100%	100	100	100	100	100	100	100	100%
11	Qwen 3.5 Plus (2026-02-15)	$0.0012	5.5s	100%	100	100	100	100	100	100	100	100%
2	Gemini 3.1 Flash Lite (Preview)	$0.0007	1.5s	100%	100	100	100	100	100	100	100	100%
9	Mistral Large 3	$0.0008	5.8s	100%	100	100	100	100	100	100	100	100%
59	GPT-4o, May 13th (temp=0)	$0.0081	3.2s	100%	100	100	100	100	100	100	100	100%
6	Gemini 3 Flash (Preview)	$0.0015	2.7s	100%	100	100	100	100	100	100	100	100%
18	Claude Haiku 4.5	$0.0027	2.2s	100%	100	100	100	100	100	100	100	100%
32	DeepSeek-V2 Chat	$0.0006	13.8s	100%	100	100	100	100	100	100	100	100%
78	ByteDance Seed 2.0 Lite	$0.0034	38.1s	100%	100	100	100	100	100	100	100	100%
49	Nemotron 3 Super	$0.0000	22.7s	100%	100	100	100	100	100	100	100	100%
85	Claude 3.5 Sonnet	$0.016	6.9s	100%	100	100	100	100	100	100	100	100%
57	GPT-4o, May 13th (temp=1)	$0.0081	2.7s	100%	100	100	100	100	100	100	100	100%
34	DeepSeek V3 (2024-12-26)	$0.0006	13.9s	100%	100	100	100	100	100	100	100	100%
61	Claude 3.7 Sonnet	$0.0080	4.6s	100%	100	100	100	100	100	100	100	100%
42	Hermes 3 405B	$0.0009	15.8s	100%	100	100	100	100	100	100	100	100%
45	GPT-4o, Aug. 6th (temp=1)	$0.0050	2.5s	100%	100	100	100	100	100	100	100	100%
43	GPT-4o, Aug. 6th (temp=0)	$0.0050	2.3s	100%	100	100	100	100	100	100	100	100%
35	Mistral Large 2	$0.0033	5.8s	100%	100	100	100	100	100	100	100	100%
1	Gemini 2.5 Flash Lite	$0.0002	1.3s	100%	100	100	100	100	100	100	100	100%
4	Gemini 2.5 Flash	$0.0012	1.8s	100%	100	100	100	100	100	100	100	100%
37	Mistral Large	$0.0033	5.9s	100%	100	100	100	100	100	100	100	100%
26	Qwen3 235B A22B Instruct 2507	$0.0003	11.9s	100%	100	100	100	100	100	100	100	100%
51	Writer: Palmyra X5	$0.0027	14.6s	100%	100	100	100	100	100	100	100	100%
8	GPT-4o Mini (temp=1)	$0.0003	7.0s	100%	100	100	100	100	100	100	100	100%
3	Mistral Small 3.2 24B	$0.0002	3.7s	100%	100	100	100	100	100	100	100	100%
12	GPT-4o Mini (temp=0)	$0.0003	8.4s	100%	100	100	100	100	100	100	100	100%
29	Llama 3.1 Nemotron 70B	$0.0011	10.8s	100%	100	100	100	100	100	100	100	100%
24	Hermes 3 70B	$0.0003	11.8s	100%	100	100	100	100	100	100	100	100%
55	WizardLM 2 8x22b	$0.0005	25.2s	100%	100	100	100	100	100	100	100	100%
82	ByteDance Seed 1.6	$0.0038	41.5s	99%	100	100	100	100	100	100	99	100%
86	o4 Mini High	$0.013	21.5s	99%	100	100	100	100	100	100	99	100%
50	MiniMax M2.7	$0.0015	17.3s	99%	100	100	100	100	100	100	99	100%
38	GPT-4.1	$0.0040	3.2s	99%	100	100	100	100	100	100	99	100%
7	Grok 4 Fast	$0.0007	4.9s	99%	100	100	100	100	100	100	99	100%
91	Z.AI GLM 4.7 Flash	$0.0014	1.3m	99%	100	100	100	100	100	100	99	100%
48	DeepSeek V3.2	$0.0003	19.0s	99%	100	100	100	100	100	100	99	100%
54	DeepSeek V3 (2025-03-24)	$0.0005	23.6s	99%	100	100	100	100	100	100	99	100%
44	Llama 3.1 70B	$0.0004	16.8s	99%	100	100	100	100	100	100	99	100%
17	ByteDance Seed 1.6 Flash	$0.0004	8.4s	99%	100	100	100	100	100	100	99	100%
62	Claude Sonnet 4.6	$0.0080	3.5s	99%	100	100	100	100	100	99	99	100%
30	GPT-5.4 Mini (Reasoning)	$0.0030	3.6s	99%	100	100	100	100	100	99	99	100%
83	Aion 2.0	$0.0034	43.1s	99%	100	100	100	100	100	99	99	100%
70	o4 Mini	$0.0085	14.6s	99%	100	100	100	100	100	99	99	100%
65	DeepSeek V3.1	$0.0005	30.2s	99%	100	100	100	100	100	100	98	100%
58	GPT-5.4 (Reasoning, Low)	$0.0072	4.6s	99%	100	100	100	100	99	99	99	100%
88	ByteDance Seed 2.0 Mini	$0.0012	1.1m	99%	100	100	100	100	99	99	99	100%
56	GPT-5.4	$0.0070	4.4s	99%	100	100	100	100	99	99	99	100%
68	GPT-5 Nano	$0.0014	33.1s	99%	100	100	100	100	99	99	99	100%
72	GPT-5 Mini	$0.0041	29.2s	99%	100	100	100	100	99	99	98	100%
46	Stealth: Hunter Alpha	$0.0000	15.8s	98%	100	100	100	100	100	100	97	100%
33	Gemma 3 27B	$0.0001	13.3s	99%	100	100	100	100	99	99	98	100%
21	Qwen 2.5 72B	$0.0002	7.7s	98%	100	100	100	99	99	99	99	100%
104	Claude Sonnet 4.6 (Reasoning)	$0.029	20.3s	99%	100	100	99	99	99	99	99	99%
84	GPT-5.4 (Reasoning)	$0.013	11.3s	98%	100	100	100	99	99	99	98	99%
13	Gemini 2.5 Flash Lite (Reasoning)	$0.0008	4.6s	99%	100	100	99	99	99	99	99	99%
5	Mistral Small 4	$0.0003	2.4s	99%	100	99	99	99	99	99	99	99%
19	GPT-4.1 Mini	$0.0008	5.1s	99%	99	99	99	99	99	98	98	99%
39	Stealth: Healer Alpha	$0.0000	7.8s	94%	100	100	100	100	100	99	91	99%
41	Mistral Small 4 (Reasoning)	$0.0010	8.2s	96%	100	100	99	99	98	98	95	99%
25	Mistral Medium 3.1	$0.0010	5.4s	98%	98	98	98	98	98	98	98	98%
10	Mistral NeMO	$0.0001	2.6s	97%	99	98	98	98	98	98	98	98%
14	GPT-5.4 Nano (Reasoning)	$0.0006	2.5s	97%	99	98	98	98	98	98	98	98%
27	Inception Mercury	$0.0003	4.9s	96%	99	99	98	98	98	97	96	98%
16	GPT-5.4 Nano (Reasoning, Low)	$0.0006	2.4s	97%	99	98	98	98	98	98	97	98%
28	Llama 3.1 8B	$0.0000	6.1s	96%	100	98	98	98	97	97	97	98%
31	Gemma 3 12B	$0.0001	7.4s	97%	99	98	98	98	98	98	98	98%
15	Ministral 3 14B	$0.0002	3.2s	97%	98	98	98	98	98	98	98	98%
23	GPT-5.4 Nano	$0.0006	2.5s	96%	98	98	98	98	98	96	96	98%
22	GPT-4.1 Nano	$0.0002	2.8s	96%	98	98	98	97	97	97	97	97%
96	Qwen 3.5 Flash	$0.0028	1.2m	89%	100	100	100	100	98	98	83	97%
53	Inception Mercury 2	$0.0012	1.6s	87%	100	100	100	99	99	99	83	97%
20	Mistral Small Creative	$0.0002	2.3s	97%	97	97	97	97	97	97	97	97%
64	Arcee AI: Trinity Mini	$0.0003	12.7s	90%	100	98	98	98	98	98	87	97%
40	Ministral 3 8B	$0.0001	3.2s	93%	98	98	98	98	96	96	91	96%
47	Ministral 8B	$0.0001	2.4s	91%	98	98	98	98	96	91	89	95%
79	Grok 4.1 Fast	$0.0006	10.5s	77%	100	100	100	100	100	100	67	95%
101	MiniMax M2.5	$0.0014	1.2m	77%	100	100	100	99	99	99	67	95%
99	Qwen 3.5 35B	$0.012	42.7s	85%	100	100	100	100	98	83	83	95%
77	Grok 4.20 (Beta)	$0.0026	1.6s	76%	100	100	100	99	99	99	66	95%
113	Qwen 3.5 9B	$0.0013	2.1m	47%	100	100	99	99	99	99	24	89%
73	GPT-5.4 Mini (Reasoning, Low)	$0.0021	3.0s	85%	91	90	89	89	88	87	84	88%
69	GPT-5.4 Mini	$0.0021	1.8s	87%	90	88	88	88	88	88	88	88%
92	Arcee AI: Trinity Large (Preview)	$0.0000	20.8s	67%	100	98	84	84	84	84	67	86%
95	Rocinante 12B	$0.0003	7.5s	57%	98	95	91	80	74	72	54	81%
87	Gemma 3 4B	$0.0001	4.7s	71%	89	78	78	78	78	77	76	79%
110	Qwen 3 32B	$0.0004	20.5s	20%	99	99	99	99	99	11	11	74%
108	Nemotron 3 Nano	$0.0004	32.3s	34%	100	98	98	92	59	41	16	72%
103	Z.AI GLM 4.5	$0.0013	11.4s	51%	100	67	67	67	67	67	67	71%
107	Ministral 3B	$0.0000	1.6s	35%	59	59	49	42	42	39	39	47%
109	Ministral 3 3B	$0.0001	1.5s	40%	42	42	41	41	40	39	39	41%
114	Cohere Command R+ (Aug. 2024)	$0.0052	18.7s	25%	57	48	33	32	32	29	18	36%
115	Claude 3 Haiku	$0.0007	3.6s	0%	98	98	0	0	0	0	0	28%
116	LFM2 24B	$0.0001	9.8s	1%	2	2	2	1	1	1	1	1%
94.91%

Median	Evaluator	Top 3	Flop 3
99.7%	Contraction expansion accuracy	100Z.AI GLM 5 Turbo 100ByteDance Seed 2.0 Lite 100Hermes 3 405B	3LFM2 24B 27Claude 3 Haiku 32Ministral 3 3B
100.0%	Non-contraction text preserved	100Hermes 3 70B 100Gemini 2.5 Flash (Reasoning) 100Aion 2.0	0LFM2 24B 29Claude 3 Haiku 40Ministral 3 3B
100.0%	Possessive traps preserved	100Grok 4 100Claude Opus 4 100Z.AI GLM 4.6	0LFM2 24B 14Z.AI GLM 4.5 29Cohere Command R+ (Aug. 2024)

Text Replacement

Expand all contractions

Performance Score Distribution (Top 20)

Price-Performance Score Distribution (Top 20)

Most Stable Models (Top 20)

Top Overall Models (Top 20)