Name replacement accuracy

Test: Text Replacement

Avg. Score

96.2%

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	Gemini 3.1 Flash Lite (Reasoning)	99.4%	$0.0009	2.6s	98%
2	Gemini 3.1 Flash Lite (Preview)	99.3%	$0.0009	1.7s	97%
3	Gemini 3.1 Flash Lite	99.4%	$0.0009	2.8s	97%
4	Grok 4 Fast	99.6%	$0.0007	5.8s	98%
5	Claude Haiku 4.5	99.7%	$0.0034	2.8s	98%
6	Grok 4.1 Fast	99.5%	$0.0008	10.6s	98%
7	DeepSeek V4 Flash	98.9%	$0.0002	7.6s	96%
8	GPT-4.1 Mini	98.9%	$0.0010	6.6s	95%
9	Gemma 4 26B	99.6%	$0.0002	15.4s	98%
10	Qwen 3.5 Plus (2026-02-15)	99.2%	$0.0015	6.7s	95%
11	Mistral Large 3	98.7%	$0.0011	7.4s	96%
12	Gemini 3 Flash (Preview)	98.4%	$0.0018	3.2s	94%
13	Stealth: Hunter Alpha	99.4%	$0.0000	17.2s	97%
14	GPT-4.1	99.4%	$0.0052	4.1s	97%
15	Inception Mercury 2	98.3%	$0.0016	2.2s	92%
16	GPT-4o Mini (temp=1)	98.2%	$0.0004	9.2s	94%
17	Qwen 2.5 72B	98.4%	$0.0003	10.3s	94%
18	Gemini 2.5 Flash Lite	98.1%	$0.0003	1.6s	88%
19	Stealth: Healer Alpha	98.9%	$0.0000	13.4s	94%
20	Grok 4.20	98.2%	$0.0019	4.2s	93%
21	GPT-4o Mini (temp=0)	98.0%	$0.0004	9.2s	93%
22	Mistral Large	98.8%	$0.0042	7.2s	96%
23	Mistral Large 2	98.8%	$0.0042	7.2s	96%
24	Gemma 3 27B	98.6%	$0.0002	16.0s	94%
25	DeepSeek V4 Pro	99.4%	$0.0012	21.5s	98%
26	GPT-4o, Aug. 6th (temp=0)	99.0%	$0.0064	2.6s	94%
27	DeepSeek-V2 Chat	98.4%	$0.0008	16.5s	94%
28	Gemini 2.5 Flash (Reasoning)	99.1%	$0.0054	10.7s	97%
29	GPT-5.4 Nano (Reasoning)	96.9%	$0.0011	5.1s	90%
30	Gemma 4 31B	99.7%	$0.0003	29.4s	99%
31	Claude Sonnet 4	99.8%	$0.010	5.7s	99%
32	GPT-4o, May 13th (temp=0)	99.7%	$0.010	3.5s	98%
33	Claude Sonnet 4.5	99.6%	$0.010	4.6s	98%
34	GPT-5.4	99.2%	$0.0089	5.4s	97%
35	Gemini 2.5 Flash	98.7%	$0.0014	2.1s	82%
36	GPT-4o, Aug. 6th (temp=1)	98.7%	$0.0064	2.6s	92%
37	GPT-5.4 (Reasoning, Low)	99.4%	$0.0094	5.6s	97%
38	Grok 4.20 (Beta)	97.6%	$0.0032	1.8s	87%
39	GPT-5.4 Nano	95.9%	$0.0007	3.1s	87%
40	Llama 3.1 8B	96.5%	$0.0000	9.3s	89%
41	Mistral Medium 3.1	96.2%	$0.0012	6.0s	89%
42	Z.AI GLM 4.5	99.7%	$0.0031	26.1s	98%
43	Claude 3.7 Sonnet	99.4%	$0.010	5.5s	97%
44	GPT-4o, May 13th (temp=1)	99.2%	$0.010	3.3s	96%
45	Claude Sonnet 4.6	99.2%	$0.010	4.4s	96%
46	GPT-5.2	98.8%	$0.0087	6.0s	94%
47	Llama 3.1 Nemotron 70B	97.7%	$0.0014	13.9s	89%
48	Mistral Small 3.2 24B	97.7%	$0.0002	4.7s	80%
49	Grok 4.20 (Reasoning)	99.6%	$0.0064	21.1s	98%
50	GPT-5.4 Mini (Reasoning)	98.4%	$0.0053	7.2s	88%
51	DeepSeek V4 Flash (Reasoning)	99.5%	$0.0005	36.8s	98%
52	Z.AI GLM 5 Turbo	99.7%	$0.0083	18.8s	98%
53	Gemma 3 12B	97.3%	$0.0001	8.6s	82%
54	Llama 3.1 70B	98.0%	$0.0005	23.9s	91%
55	GPT-5.4 Nano (Reasoning, Low)	95.3%	$0.0007	3.7s	83%
56	Gemini 3 Flash (Preview, Reasoning)	99.5%	$0.0097	17.2s	98%
57	GPT-5.4 Mini	96.1%	$0.0027	2.1s	82%
58	Qwen 3.6 35B	99.4%	$0.0055	29.6s	98%
59	Grok 4.20 (Beta, Reasoning)	99.5%	$0.014	7.8s	98%
60	GPT-5.4 Mini (Reasoning, Low)	96.1%	$0.0029	3.3s	82%
61	Grok 4.3	96.3%	$0.0020	4.5s	80%
62	Xiaomi MIMO v2.5	98.2%	$0.0033	13.1s	83%
63	GPT-5.1	99.5%	$0.013	13.7s	98%
64	GPT-5 Mini	99.3%	$0.0052	33.1s	97%
65	GPT-5.5	99.7%	$0.018	4.7s	99%
66	DeepSeek V3 (2024-12-26)	97.6%	$0.0007	15.0s	80%
67	DeepSeek V3.2	98.9%	$0.0004	44.5s	96%
68	Claude Opus 4.6	99.5%	$0.017	5.6s	98%
69	Z.AI GLM 4.5 Air	98.3%	$0.0015	30.2s	89%
70	Claude Opus 4.5	99.3%	$0.017	5.2s	97%
71	Z.AI GLM 4.6	99.7%	$0.0045	42.4s	98%
72	GPT-OSS 120B	98.9%	$0.0008	48.6s	97%
73	ByteDance Seed 2.0 Lite	99.7%	$0.0040	44.8s	98%
74	Mistral Small 4	95.2%	$0.0004	3.2s	72%
75	ByteDance Seed 1.6	99.4%	$0.0040	43.2s	98%
76	Inception Mercury	94.0%	$0.0004	4.5s	76%
77	GPT-5.4 (Reasoning)	99.2%	$0.016	13.3s	97%
78	ByteDance Seed 1.6 Flash	95.5%	$0.0007	12.2s	76%
79	Grok 4.3 (Reasoning)	99.3%	$0.0079	35.9s	97%
80	GPT-5.5 (Reasoning, Low)	99.7%	$0.021	6.7s	99%
81	Aion 2.0	99.5%	$0.0038	48.1s	97%
82	Qwen3 235B A22B Instruct 2507	95.6%	$0.0003	14.8s	75%
83	GPT-5 Nano	98.8%	$0.0023	50.9s	96%
84	Claude 3.5 Sonnet	99.3%	$0.020	9.0s	97%
85	Ministral 3 14B	93.4%	$0.0002	3.9s	70%
86	Qwen 3.5 Flash	98.7%	$0.0027	48.6s	93%
87	Mistral Small Creative	92.7%	$0.0002	2.9s	71%
88	Hermes 3 405B	97.3%	$0.0012	22.3s	74%
89	Xiaomi MIMO v2.5 Pro	97.4%	$0.0035	15.5s	72%
90	Gemini 2.5 Flash Lite (Reasoning)	96.3%	$0.0018	16.2s	71%
91	Qwen 3.6 Flash	98.2%	$0.0074	21.9s	80%
92	Claude Opus 4.7	98.5%	$0.023	4.3s	95%
93	Writer: Palmyra X5	94.3%	$0.0034	11.1s	73%
94	Claude Opus 4.7 (Reasoning)	98.7%	$0.025	4.7s	95%
95	Mistral Small 4 (Reasoning)	94.1%	$0.0017	14.8s	69%
96	GPT-5.5 (Reasoning)	99.4%	$0.027	9.8s	97%
97	Qwen 3.5 35B	99.0%	$0.013	43.1s	96%
98	Z.AI GLM 5	99.5%	$0.0083	60.0s	97%
99	Qwen 3 32B	95.3%	$0.0008	32.0s	72%
100	o4 Mini	97.7%	$0.013	21.3s	80%
101	Grok 4	99.6%	$0.023	28.7s	98%
102	Gemini 2.5 Pro	99.6%	$0.027	19.7s	98%
103	WizardLM 2 8x22b	96.2%	$0.0007	37.2s	70%
104	o4 Mini High	99.1%	$0.021	33.0s	96%
105	Claude Opus 4.6 (Reasoning)	99.6%	$0.030	12.5s	98%
106	MiniMax M2.5	97.7%	$0.0018	59.0s	80%
107	Arcee AI: Trinity Mini	88.5%	$0.0002	6.5s	58%
108	GPT-4.1 Nano	89.9%	$0.0003	3.6s	52%
109	Qwen 3.5 Plus (2026-04-20)	99.4%	$0.011	1.2m	98%
110	Z.AI GLM 5.1	99.5%	$0.013	1.1m	98%
111	Claude Sonnet 4.6 (Reasoning)	99.6%	$0.032	19.3s	98%
112	Qwen 3.5 27B	99.5%	$0.015	1.1m	98%
113	Z.AI GLM 4.7	99.7%	$0.0070	1.5m	97%
114	Nemotron 3 Super	95.2%	$0.0000	51.1s	66%
115	Ministral 3 8B	85.7%	$0.0002	3.3s	52%
116	Mistral NeMO	85.5%	$0.0002	2.5s	51%
117	DeepSeek V4 Pro (Reasoning)	99.7%	$0.0064	1.7m	98%
118	DeepSeek V3 (2025-03-24)	92.5%	$0.0006	35.4s	58%
119	Gemma 4 31B (Reasoning)	99.4%	$0.0011	1.9m	98%
120	GPT-5	99.8%	$0.029	44.3s	99%
121	Gemma 4 26B (Reasoning)	99.5%	$0.0013	1.9m	98%
122	Ministral 8B	84.8%	$0.0001	3.3s	50%
123	Qwen 3.5 122B	99.6%	$0.025	59.3s	98%
124	ByteDance Seed 2.0 Mini	97.6%	$0.0017	1.7m	91%
125	Qwen 3.6 27B	97.6%	$0.014	58.6s	81%
126	Gemini 3 Pro (Preview)	99.5%	$0.038	26.2s	98%
127	MiniMax M2.7	96.7%	$0.0058	1.2m	74%
128	Z.AI GLM 4.7 Flash	94.0%	$0.0015	1.1m	68%
129	Claude Opus 4	99.8%	$0.051	7.7s	98%
130	Arcee AI: Trinity Large (Preview)	88.0%	$0.0000	23.8s	42%
131	DeepSeek V3.1	88.5%	$0.0006	34.3s	46%
132	Qwen 3.5 9B	96.8%	$0.0012	1.8m	78%
133	Ministral 3 3B	78.9%	$0.0001	2.0s	43%
134	MoonshotAI: Kimi K2.5	99.6%	$0.011	2.1m	98%
135	Ministral 3B	78.4%	$0.0000	2.0s	39%
136	Qwen 3.5 397B A17B	99.4%	$0.0089	2.2m	97%
137	MoonshotAI: Kimi K2.6	99.6%	$0.018	1.9m	98%
138	Gemini 3.1 Pro (Preview)	98.8%	$0.045	42.6s	87%
139	LFM2 24B	76.9%	$0.0001	11.9s	30%
140	Qwen3.6 Max Preview	99.7%	$0.033	1.8m	99%
141	Gemma 3 4B	73.9%	$0.0001	6.0s	24%
142	Rocinante 12B	63.8%	$0.0004	8.0s	17%
143	Cohere Command R+ (Aug. 2024)	71.4%	$0.0067	27.2s	23%
144	Nemotron 3 Nano	83.4%	$0.0016	2.0m	46%
145	Claude 3 Haiku	61.2%	$0.0009	4.3s	4%
146	Hermes 3 70B	67.7%	$0.0015	2.0m	10%
96.23%

Individual Scenarios

Generic Prompt

▼

Character rename: Elena->Mirabel, Gregor->Aldric

Model	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen3.6 Max Preview	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5.1	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	100	100	100	100.0%
Grok 4.3 (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100	100	100.0%
Claude Opus 4.7 (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100	100	100.0%
Gemma 4 31B (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-04-20)	100	100	100	100	100	100	100	100.0%
Gemma 4 26B (Reasoning)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Reasoning)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100	100	100.0%
Qwen 3.6 Flash	100	100	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Pro (Reasoning)	100	100	100	100	100	100	100	100.0%
Claude Opus 4.7	100	100	100	100	100	100	100	100.0%
Qwen 3.6 27B	100	100	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100	100	100.0%
Aion 2.0	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100	100	100.0%
MiniMax M2.7	100	100	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100	100	100.0%
Qwen 3.6 35B	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Flash (Reasoning)	100	100	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100	100	100.0%
MiniMax M2.5	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 35B	100	100	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100	100	100.0%
Xiaomi MIMO v2.5 Pro	100	100	100	100	100	100	100	100.0%
Stealth: Hunter Alpha	100	100	100	100	100	100	100	100.0%
Gemma 4 31B	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-OSS 120B	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100	100	100.0%
Stealth: Healer Alpha	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	100	100	100	100.0%
Gemma 4 26B	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite	100	100	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100	100	100.0%
Claude Haiku 4.5	100	100	100	100	100	100	100	100.0%
Xiaomi MIMO v2.5	100	100	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Beta)	100	100	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.5 Air	100	100	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Pro	100	100	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100	100	100.0%
Mistral Small 4 (Reasoning)	100	100	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Flash	100	100	100	100	100	100	100	100.0%
Grok 4.20	100	100	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning)	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100	100	100.0%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100	100	100.0%
Grok 4.3	100	100	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100	100	100.0%
Nemotron 3 Nano	100	100	100	100	100	100	100	100.0%
Mistral Small 4	100	100	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100	100	100.0%
GPT-5.4 Nano	100	100	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100	100	100.0%
Ministral 3 14B	100	100	100	100	100	100	100	100.0%
GPT-4.1 Nano	100	100	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100	100	100.0%
LFM2 24B	100	100	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Mini	100	100	100	100	100	100	92	98.8%
Qwen 3 32B	100	100	100	100	100	100	75	96.4%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100	75	96.4%
Rocinante 12B	100	100	100	100	92	83	75	92.9%
Ministral 3 3B	92	92	92	92	92	92	92	91.7%
Ministral 3B	100	100	92	92	92	92	58	89.3%
Mistral NeMO	100	100	100	100	100	100	8	86.9%
DeepSeek V3.1	100	100	100	100	100	92	8	85.7%
Arcee AI: Trinity Mini	100	100	92	92	92	83	0	79.8%
Cohere Command R+ (Aug. 2024)	100	100	8	0	0	0	0	29.8%
Claude 3 Haiku	100	100	0	0	0	0	0	28.6%

▼

Combined: 3rd person past → 1st person present

Model	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Avg ▼
Grok 4.1 Fast	100	100	100	100	100	100	98	99.7%
Z.AI GLM 5 Turbo	100	100	100	100	100	98	98	99.5%
Grok 4 Fast	100	100	100	100	100	98	98	99.5%
GPT-5.5	100	100	100	100	98	98	98	99.2%
Z.AI GLM 4.5	100	100	100	100	98	98	98	99.2%
GPT-5.4 (Reasoning, Low)	100	100	100	98	98	98	96	98.7%
GPT-5.2	100	100	100	98	98	98	96	98.7%
GPT-5.5 (Reasoning)	100	100	98	98	98	98	98	98.7%
ByteDance Seed 1.6	100	100	98	98	98	98	98	98.7%
GPT-5 Nano	100	100	100	98	98	98	96	98.7%
Grok 4.3 (Reasoning)	100	100	100	98	98	98	95	98.5%
Qwen 3.5 Plus (2026-04-20)	100	100	98	98	98	98	96	98.5%
Claude Opus 4	100	100	100	98	98	96	96	98.5%
GPT-5.5 (Reasoning, Low)	100	98	98	98	98	98	98	98.5%
Z.AI GLM 4.7	100	98	98	98	98	98	98	98.5%
GPT-5.4 Mini	100	98	98	98	98	98	96	98.2%
Claude Opus 4.6 (Reasoning)	98	98	98	98	98	98	98	98.2%
Qwen3.6 Max Preview	98	98	98	98	98	98	98	98.2%
GPT-5.4 (Reasoning)	100	98	98	98	98	98	96	98.2%
Claude Opus 4.7 (Reasoning)	98	98	98	98	98	98	98	98.2%
Claude Opus 4.6	98	98	98	98	98	98	98	98.2%
GPT-5	100	98	98	98	98	98	96	98.2%
Gemma 4 31B (Reasoning)	98	98	98	98	98	98	98	98.2%
GPT-5.4 Mini (Reasoning)	98	98	98	98	98	98	98	98.2%
Claude Sonnet 4	98	98	98	98	98	98	98	98.2%
Grok 4	100	98	98	98	98	98	96	98.2%
Gemma 4 31B	98	98	98	98	98	98	98	98.2%
Qwen 3.5 122B	98	98	98	98	98	98	96	98.0%
Grok 4.20 (Beta, Reasoning)	98	98	98	98	98	98	96	98.0%
Grok 4.20 (Reasoning)	98	98	98	98	98	98	96	98.0%
Qwen 3.5 27B	98	98	98	98	98	98	96	98.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	98	98	96	93	98.0%
Z.AI GLM 4.6	98	98	98	98	98	98	96	98.0%
DeepSeek V4 Pro (Reasoning)	100	100	98	98	96	96	95	97.7%
Claude Haiku 4.5	100	98	98	98	96	96	96	97.7%
Aion 2.0	98	98	98	98	98	96	96	97.7%
MiniMax M2.5	98	98	98	98	98	96	96	97.7%
Claude Sonnet 4.6 (Reasoning)	100	98	98	96	96	96	96	97.5%
GPT-5.1	98	98	98	98	98	96	95	97.5%
DeepSeek V4 Flash (Reasoning)	100	98	98	96	96	96	96	97.5%
Gemini 2.5 Pro	98	98	98	98	96	96	96	97.5%
GPT-5.4	98	98	98	98	96	96	96	97.5%
MoonshotAI: Kimi K2.5	100	98	96	96	96	96	96	97.2%
Claude Opus 4.7	98	98	98	98	96	96	95	97.2%
Gemini 3 Pro (Preview)	98	98	98	96	96	96	96	97.2%
Claude Sonnet 4.5	98	98	98	96	96	96	96	97.2%
DeepSeek V4 Pro	98	98	98	96	96	96	96	97.2%
Qwen 3.6 35B	98	98	98	96	96	96	96	97.2%
GPT-OSS 120B	98	98	98	96	96	96	95	97.0%
DeepSeek V3.1	98	98	96	96	96	96	96	97.0%
Gemma 4 26B (Reasoning)	98	98	96	96	96	96	95	96.7%
Qwen 3.5 35B	98	98	96	96	96	96	95	96.7%
Stealth: Hunter Alpha	98	96	96	96	96	96	96	96.7%
Qwen 3.5 Flash	98	96	96	96	96	96	96	96.7%
Gemma 4 26B	98	96	96	96	96	96	96	96.7%
GPT-4.1 Mini	98	98	98	96	96	96	93	96.7%
Z.AI GLM 4.5 Air	98	96	96	96	96	96	96	96.7%
Hermes 3 405B	98	96	96	96	96	96	96	96.7%
Xiaomi MIMO v2.5 Pro	98	98	96	96	96	96	95	96.7%
Gemini 3.1 Pro (Preview)	96	96	96	96	96	96	96	96.5%
Z.AI GLM 5.1	96	96	96	96	96	96	96	96.5%
MoonshotAI: Kimi K2.6	96	96	96	96	96	96	96	96.5%
Qwen 3.5 397B A17B	96	96	96	96	96	96	96	96.5%
Claude Sonnet 4.6	96	96	96	96	96	96	96	96.5%
Qwen 3.6 Flash	98	96	96	96	96	96	95	96.5%
Claude Opus 4.5	96	96	96	96	96	96	96	96.5%
MiniMax M2.7	98	98	96	96	96	96	93	96.5%
Qwen 3.5 Plus (2026-02-15)	96	96	96	96	96	96	96	96.5%
GPT-5.4 Mini (Reasoning, Low)	98	98	96	96	96	95	95	96.5%
GPT-4o, May 13th (temp=0)	96	96	96	96	96	96	96	96.5%
ByteDance Seed 2.0 Lite	96	96	96	96	96	96	96	96.5%
Claude 3.5 Sonnet	96	96	96	96	96	96	96	96.5%
Claude 3.7 Sonnet	96	96	96	96	96	96	96	96.5%
GPT-4o, Aug. 6th (temp=0)	96	96	96	96	96	96	96	96.5%
DeepSeek V3.2	96	96	96	96	96	96	96	96.5%
GPT-4o Mini (temp=1)	96	96	96	96	96	96	96	96.5%
GPT-4o Mini (temp=0)	96	96	96	96	96	96	96	96.5%
Gemini 3 Flash (Preview, Reasoning)	98	96	96	96	96	96	93	96.2%
Gemini 3.1 Flash Lite (Reasoning)	96	96	96	96	96	96	95	96.2%
GPT-5.4 Nano (Reasoning, Low)	96	96	96	96	96	96	95	96.2%
GPT-5.4 Nano	96	96	96	96	96	96	95	96.2%
Hermes 3 70B	96	96	96	96	96	96	95	96.2%
GPT-5 Mini	98	98	96	96	96	95	93	96.2%
Gemma 3 27B	98	96	96	96	95	95	95	96.0%
Z.AI GLM 5	96	96	96	96	96	96	93	96.0%
Xiaomi MIMO v2.5	100	96	96	96	96	95	91	96.0%
GPT-4.1	96	96	96	96	95	95	95	95.7%
Stealth: Healer Alpha	98	96	96	96	96	93	93	95.7%
Gemini 2.5 Flash Lite	96	96	96	96	95	95	95	95.7%
Gemini 3.1 Flash Lite	96	96	96	96	95	95	93	95.5%
Z.AI GLM 4.7 Flash	98	96	96	95	95	95	93	95.5%
GPT-5.4 Nano (Reasoning)	96	96	96	96	96	95	91	95.5%
Qwen 2.5 72B	96	96	96	95	95	95	95	95.5%
GPT-4o, May 13th (temp=1)	98	96	96	95	95	93	93	95.2%
Gemini 3 Flash (Preview)	96	96	96	95	95	95	93	95.2%
DeepSeek V4 Flash	96	96	96	96	96	95	88	95.0%
WizardLM 2 8x22b	100	98	96	96	93	91	89	95.0%
Gemini 2.5 Flash (Reasoning)	96	96	95	95	95	93	93	94.7%
Gemini 3.1 Flash Lite (Preview)	95	95	95	95	95	95	95	94.7%
GPT-4o, Aug. 6th (temp=1)	98	96	95	95	95	91	91	94.5%
o4 Mini High	98	96	95	95	95	91	89	94.2%
Inception Mercury	98	96	96	95	93	93	82	93.5%
Mistral Small 3.2 24B	95	95	93	93	93	93	93	93.5%
GPT-4.1 Nano	96	95	95	93	93	91	89	93.2%
Mistral Large 3	93	93	93	93	93	93	93	93.0%
Mistral Large 2	93	93	93	93	93	93	93	93.0%
Mistral Large	93	93	93	93	93	93	93	93.0%
Inception Mercury 2	100	98	98	96	96	95	63	92.5%
DeepSeek V3 (2024-12-26)	93	93	93	93	93	91	89	92.2%
o4 Mini	95	95	93	93	91	89	88	92.0%
DeepSeek-V2 Chat	93	93	93	91	91	91	91	92.0%
LFM2 24B	93	93	93	93	91	89	89	91.7%
DeepSeek V3 (2025-03-24)	93	93	91	91	91	91	89	91.5%
Qwen 3.5 9B	98	98	98	96	96	96	53	91.0%
Grok 4.20 (Beta)	96	91	91	89	89	88	88	90.5%
Mistral Medium 3.1	91	91	88	88	88	88	88	88.7%
ByteDance Seed 2.0 Mini	95	93	93	89	86	82	75	87.7%
Grok 4.20	91	89	89	88	88	86	79	87.2%
Grok 4.3	100	100	98	98	98	67	49	87.2%
Qwen 3 32B	96	93	93	91	84	77	68	86.2%
Llama 3.1 70B	93	89	84	84	84	82	81	85.5%
Gemini 2.5 Flash	98	98	98	98	98	95	11	85.2%
Llama 3.1 Nemotron 70B	86	86	86	86	84	82	82	84.7%
Qwen 3.6 27B	100	98	98	98	98	72	26	84.5%
Mistral Small Creative	93	82	82	82	82	82	81	83.7%
Mistral Small 4 (Reasoning)	98	96	93	93	93	89	23	83.7%
Gemma 3 12B	96	96	95	95	95	95	11	83.2%
Llama 3.1 8B	89	88	88	84	79	77	75	83.0%
ByteDance Seed 1.6 Flash	93	93	88	84	81	81	26	77.9%
Mistral Small 4	95	93	93	93	89	40	30	76.2%
Nemotron 3 Super	100	100	100	98	96	18	11	74.7%
Arcee AI: Trinity Large (Preview)	96	96	96	96	96	11	11	71.9%
Nemotron 3 Nano	98	98	96	93	89	14	12	71.7%
Ministral 3 3B	89	84	70	70	67	63	58	71.7%
Mistral NeMO	77	75	72	70	70	68	67	71.4%
Qwen3 235B A22B Instruct 2507	82	81	79	67	65	60	58	70.2%
Ministral 3B	89	89	79	63	58	49	47	67.9%
Writer: Palmyra X5	74	67	65	61	61	60	58	63.7%
Ministral 3 14B	63	63	63	63	61	58	56	61.2%
Arcee AI: Trinity Mini	96	82	79	79	77	7	4	60.7%
Ministral 8B	53	51	47	44	44	44	42	46.4%
Ministral 3 8B	54	49	46	44	44	42	40	45.6%
Claude 3 Haiku	93	89	0	0	0	0	0	26.1%
Cohere Command R+ (Aug. 2024)	89	16	12	4	2	0	0	17.5%
Rocinante 12B	93	12	9	4	2	0	0	17.0%
Gemma 3 4B	11	11	11	11	11	11	11	10.5%

▼

Expand all contractions

Model	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen3.6 Max Preview	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5.1	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100	100	100.0%
Gemma 4 26B (Reasoning)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Reasoning)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100	100	100.0%
Qwen 3.6 27B	100	100	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100	100	100.0%
Qwen 3.6 35B	100	100	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100	100	100.0%
Gemma 4 31B	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Reasoning)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	100	100	100	100.0%
Gemma 4 26B	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite	100	100	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100	100	100.0%
Claude Haiku 4.5	100	100	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.5 Air	100	100	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Flash	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100	100	100.0%
Grok 4.3	100	100	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100	100	100.0%
Hermes 3 70B	100	100	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100	100	100.0%
Grok 4.3 (Reasoning)	100	100	100	100	100	100	98	99.7%
ByteDance Seed 1.6	100	100	100	100	100	100	98	99.7%
Qwen 3.6 Flash	100	100	100	100	100	100	98	99.7%
o4 Mini High	100	100	100	100	100	100	98	99.7%
MiniMax M2.7	100	100	100	100	100	100	98	99.7%
DeepSeek V4 Flash (Reasoning)	100	100	100	100	100	100	98	99.7%
GPT-4.1	100	100	100	100	100	100	98	99.7%
GPT-OSS 120B	100	100	100	100	100	100	98	99.7%
Grok 4 Fast	100	100	100	100	100	100	98	99.7%
Xiaomi MIMO v2.5	100	100	100	100	100	100	98	99.7%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100	98	99.7%
DeepSeek V3.2	100	100	100	100	100	100	98	99.7%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100	98	99.7%
Llama 3.1 70B	100	100	100	100	100	100	98	99.7%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100	98	99.7%
Qwen 3.5 Plus (2026-04-20)	100	100	100	100	100	98	98	99.3%
Claude Sonnet 4.6	100	100	100	100	100	98	98	99.3%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	98	98	99.3%
DeepSeek V4 Pro (Reasoning)	100	100	100	100	100	100	95	99.3%
Aion 2.0	100	100	100	100	100	98	98	99.3%
o4 Mini	100	100	100	100	100	98	98	99.3%
Qwen 3.5 35B	100	100	100	100	100	100	95	99.3%
DeepSeek V3.1	100	100	100	100	100	100	95	99.3%
GPT-5.4 (Reasoning, Low)	100	100	100	100	98	98	98	99.0%
Claude Opus 4.7	100	100	100	100	98	98	98	99.0%
ByteDance Seed 2.0 Mini	100	100	100	100	98	98	98	99.0%
GPT-5.4	100	100	100	100	98	98	98	99.0%
GPT-5 Nano	100	100	100	100	98	98	98	99.0%
GPT-5 Mini	100	100	100	100	98	98	95	98.7%
Xiaomi MIMO v2.5 Pro	100	100	100	100	98	98	95	98.7%
Stealth: Hunter Alpha	100	100	100	100	100	100	91	98.7%
Qwen 3.5 Flash	100	100	100	100	100	95	95	98.7%
Grok 4.20 (Beta)	100	100	100	98	98	98	98	98.7%
Inception Mercury 2	100	100	100	98	98	98	98	98.7%
Gemma 3 27B	100	100	100	100	98	98	95	98.7%
Qwen 2.5 72B	100	100	100	98	98	98	98	98.7%
Claude Sonnet 4.6 (Reasoning)	100	100	98	98	98	98	98	98.3%
GPT-5.4 (Reasoning)	100	100	100	98	98	98	95	98.3%
Gemini 2.5 Flash Lite (Reasoning)	100	100	98	98	98	98	98	98.3%
DeepSeek V4 Pro	100	100	98	98	98	98	98	98.3%
Grok 4.20	100	98	98	98	98	98	98	98.0%
Mistral Small 4	100	98	98	98	98	98	98	98.0%
Claude Opus 4.7 (Reasoning)	98	98	98	98	98	98	98	97.7%
Gemma 4 31B (Reasoning)	100	98	98	98	98	98	95	97.7%
GPT-4.1 Mini	98	98	98	98	98	95	95	97.0%
Stealth: Healer Alpha	100	100	100	100	100	98	74	96.0%
Mistral Small 4 (Reasoning)	100	100	98	98	95	95	86	96.0%
Mistral Medium 3.1	95	95	95	95	95	95	95	95.3%
Mistral NeMO	98	95	95	95	95	93	93	95.0%
GPT-5.4 Nano (Reasoning)	98	95	95	95	93	93	93	94.7%
Inception Mercury	98	98	95	95	95	91	88	94.4%
GPT-5.4 Nano (Reasoning, Low)	98	95	95	95	93	93	91	94.4%
Llama 3.1 8B	100	95	95	95	91	91	91	94.0%
Gemma 3 12B	98	93	93	93	93	93	93	93.7%
Ministral 3 14B	95	95	93	93	93	93	93	93.7%
GPT-5.4 Nano	95	95	95	95	93	88	88	93.0%
GPT-4.1 Nano	93	93	93	91	91	91	91	91.7%
Mistral Small Creative	91	91	91	91	91	91	91	90.7%
Arcee AI: Trinity Mini	100	95	95	93	93	93	60	90.0%
Qwen 3.5 9B	100	100	98	98	98	98	33	89.0%
Ministral 3 8B	93	93	93	93	88	88	72	88.7%
Ministral 8B	95	93	93	93	88	72	67	86.0%
MiniMax M2.5	100	100	100	98	98	98	2	85.0%
Qwen 3 32B	98	98	98	98	98	12	12	73.1%
Nemotron 3 Nano	100	95	95	95	47	44	28	72.1%
GPT-5.4 Mini (Reasoning, Low)	74	70	67	67	65	60	51	65.1%
GPT-5.4 Mini	70	65	65	65	63	63	63	64.8%
Arcee AI: Trinity Large (Preview)	100	95	53	51	51	51	2	57.8%
Rocinante 12B	95	86	74	40	21	16	2	47.8%
Gemma 3 4B	67	35	35	33	33	30	28	37.2%
Ministral 3B	37	37	37	37	37	28	26	34.2%
Ministral 3 3B	37	37	33	33	30	28	28	32.2%
Cohere Command R+ (Aug. 2024)	60	37	37	33	28	16	14	32.2%
Claude 3 Haiku	93	93	0	0	0	0	0	26.6%
LFM2 24B	5	5	5	2	2	2	2	3.3%

▼

Location rename: market square, outer ring, bridge, northern mines

Model	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen3.6 Max Preview	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5.1	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	100	100	100	100.0%
Grok 4.3 (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100	100	100.0%
Gemma 4 31B (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen 3.5 122B	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-04-20)	100	100	100	100	100	100	100	100.0%
Gemma 4 26B (Reasoning)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Reasoning)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 27B	100	100	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100	100	100.0%
Qwen 3.6 Flash	100	100	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Pro (Reasoning)	100	100	100	100	100	100	100	100.0%
Claude Opus 4.7	100	100	100	100	100	100	100	100.0%
Qwen 3.6 27B	100	100	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100	100	100.0%
Aion 2.0	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100	100	100.0%
Qwen 3.6 35B	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Flash (Reasoning)	100	100	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100	100	100.0%
MiniMax M2.5	100	100	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100	100	100.0%
Qwen 3.5 35B	100	100	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100	100	100.0%
Xiaomi MIMO v2.5 Pro	100	100	100	100	100	100	100	100.0%
Stealth: Hunter Alpha	100	100	100	100	100	100	100	100.0%
Gemma 4 31B	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-OSS 120B	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Flash	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100	100	100.0%
Qwen 3.5 9B	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100	100	100.0%
Stealth: Healer Alpha	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	100	100	100	100.0%
Gemma 4 26B	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite	100	100	100	100	100	100	100	100.0%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite (Reasoning)	100	100	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100	100	100.0%
Claude Haiku 4.5	100	100	100	100	100	100	100	100.0%
Xiaomi MIMO v2.5	100	100	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100	100	100.0%
Nemotron 3 Super	100	100	100	100	100	100	100	100.0%
GPT-5.4	100	100	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100	100	100.0%
Inception Mercury 2	100	100	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.5 Air	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Pro	100	100	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100	100	100.0%
GPT-5.4 Mini	100	100	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Flash	100	100	100	100	100	100	100	100.0%
Grok 4.20	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100	100	100.0%
Inception Mercury	100	100	100	100	100	100	100	100.0%
Grok 4.3	100	100	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning)	100	100	100	100	100	100	88	98.2%
MiniMax M2.7	100	100	100	100	100	100	88	98.2%
Z.AI GLM 4.7	100	100	100	100	100	100	88	98.2%
DeepSeek-V2 Chat	100	100	100	100	100	100	88	98.2%
Grok 4.20 (Beta)	100	100	100	100	100	100	88	98.2%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100	88	98.2%
DeepSeek V3.1	100	100	100	100	100	100	88	98.2%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100	88	98.2%
Claude Sonnet 4.6	100	100	100	100	100	88	88	96.4%
Z.AI GLM 4.7 Flash	100	100	100	100	100	88	88	96.4%
Hermes 3 405B	100	100	100	100	100	88	88	96.4%
Mistral Small 4 (Reasoning)	100	100	100	100	100	88	88	96.4%
Qwen 3 32B	100	100	100	100	100	88	88	96.4%
Nemotron 3 Nano	100	100	100	100	100	88	88	96.4%
Mistral Small 4	100	100	100	100	100	88	88	96.4%
Llama 3.1 8B	100	100	100	100	100	100	75	96.4%
Claude Opus 4.7 (Reasoning)	100	100	100	100	88	88	88	94.6%
ByteDance Seed 2.0 Mini	100	100	100	100	88	88	88	94.6%
Gemma 3 27B	100	100	100	100	88	88	88	94.6%
Arcee AI: Trinity Large (Preview)	100	100	100	100	88	88	88	94.6%
Hermes 3 70B	100	100	100	100	88	88	88	94.6%
GPT-5.4 Nano (Reasoning, Low)	100	100	100	88	88	88	88	92.9%
GPT-4.1 Nano	100	100	100	100	100	88	63	92.9%
Gemini 3 Flash (Preview)	100	100	88	88	88	88	88	91.1%
GPT-4o Mini (temp=1)	100	100	88	88	88	88	88	91.1%
Qwen 2.5 72B	100	100	88	88	88	88	88	91.1%
GPT-5.4 Nano	100	100	88	88	88	88	88	91.1%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	88	38	89.3%
GPT-5.4 Nano (Reasoning)	100	88	88	88	88	88	88	89.3%
GPT-4o Mini (temp=0)	88	88	88	88	88	88	88	87.5%
Mistral Medium 3.1	88	88	88	88	88	88	88	87.5%
Mistral Small Creative	88	88	88	88	88	88	88	87.5%
Ministral 3 14B	88	88	88	88	88	88	88	87.5%
Arcee AI: Trinity Mini	88	88	88	88	88	88	88	87.5%
Gemma 3 4B	88	88	88	88	88	88	88	87.5%
Ministral 3 3B	63	50	50	50	38	38	38	46.4%
Ministral 3B	88	63	50	38	38	25	25	46.4%
Mistral NeMO	75	50	38	38	38	25	25	41.1%
Ministral 3 8B	38	38	38	38	38	38	25	35.7%
Cohere Command R+ (Aug. 2024)	75	50	50	25	25	13	13	35.7%
Rocinante 12B	88	50	50	25	13	13	0	33.9%
Ministral 8B	38	38	38	38	25	25	13	30.4%
LFM2 24B	25	25	25	25	25	25	25	25.0%
Claude 3 Haiku	0	0	0	0	0	0	0	0.0%

▼

Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged

Model	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Avg ▼
Claude Opus 4.6 (Reasoning)	100	100	100	100	100	100	100	100.0%
Qwen3.6 Max Preview	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.6 (Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.5 (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.6	100	100	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100	100	100.0%
Gemma 4 26B (Reasoning)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Beta, Reasoning)	100	100	100	100	100	100	100	100.0%
GPT-5.4 (Reasoning, Low)	100	100	100	100	100	100	100	100.0%
Grok 4.20 (Reasoning)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview, Reasoning)	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Pro (Reasoning)	100	100	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100	100	100.0%
Aion 2.0	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100	100	100.0%
MiniMax M2.7	100	100	100	100	100	100	100	100.0%
GPT-5.5	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Flash (Reasoning)	100	100	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100	100	100.0%
Stealth: Hunter Alpha	100	100	100	100	100	100	100	100.0%
Gemma 4 31B	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Reasoning)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite (Preview)	100	100	100	100	100	100	100	100.0%
Gemma 4 26B	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Flash Lite	100	100	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100	100	100.0%
Claude Haiku 4.5	100	100	100	100	100	100	100	100.0%
ByteDance Seed 2.0 Lite	100	100	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100	100	100.0%
GPT-4.1 Mini	100	100	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100	100	100.0%
DeepSeek V4 Pro	100	100	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100	100	100.0%
Qwen 3 32B	100	100	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100	100	100.0%
Qwen3 235B A22B Instruct 2507	100	100	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5.1	100	100	100	100	100	100	97	99.6%
Claude Opus 4.7 (Reasoning)	100	100	100	100	100	100	97	99.6%
GPT-5.5 (Reasoning)	100	100	100	100	100	100	97	99.6%
o4 Mini High	100	100	100	100	100	100	97	99.6%
Mistral Medium 3.1	100	100	100	100	100	100	97	99.6%
Z.AI GLM 5 Turbo	100	100	100	100	100	100	95	99.2%
Qwen 3.5 122B	100	100	100	100	100	97	97	99.2%
Qwen 3.6 27B	100	100	100	100	100	97	97	99.2%
o4 Mini	100	100	100	100	100	97	97	99.2%
Z.AI GLM 4.5 Air	100	100	100	100	100	97	97	99.2%
Writer: Palmyra X5	100	100	100	100	100	100	95	99.2%
Llama 3.1 70B	100	100	100	100	100	100	95	99.2%
Claude 3 Haiku	100	100	100	100	100	100	95	99.2%
Arcee AI: Trinity Mini	100	100	100	100	100	97	97	99.2%
Ministral 3 3B	100	100	100	100	100	97	97	99.2%
GPT-5.4 (Reasoning)	100	100	100	100	100	97	95	98.8%
Grok 4.3 (Reasoning)	100	100	100	100	97	97	97	98.8%
GPT-5.4	100	100	100	100	97	97	97	98.8%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	97	97	97	98.8%
Qwen 3.5 397B A17B	100	100	100	100	100	95	95	98.5%
GPT-5.2	100	100	100	100	97	97	95	98.5%
Qwen 3.6 35B	100	100	100	97	97	97	97	98.5%
Grok 4.20	100	100	100	100	100	95	95	98.5%
Gemma 4 31B (Reasoning)	100	100	100	97	97	97	97	98.5%
GPT-5.4 Mini (Reasoning)	100	100	100	100	100	95	95	98.5%
Xiaomi MIMO v2.5	100	100	100	100	97	97	95	98.5%
GPT-5 Mini	100	100	100	97	97	97	95	98.1%
Qwen 3.5 Plus (2026-04-20)	100	100	100	97	97	97	95	98.1%
Qwen 3.5 9B	100	100	100	97	97	97	95	98.1%
Ministral 3B	100	100	100	97	97	97	95	98.1%
GPT-5.4 Mini (Reasoning, Low)	100	100	100	100	95	95	95	97.7%
Qwen 3.6 Flash	100	100	97	97	97	97	95	97.7%
Stealth: Healer Alpha	100	97	97	97	97	97	97	97.7%
Qwen 2.5 72B	100	100	100	100	95	95	95	97.7%
Inception Mercury	100	100	100	97	97	95	95	97.7%
Qwen 3.5 27B	97	97	97	97	97	97	97	97.3%
MiniMax M2.5	100	97	97	97	97	97	95	97.3%
Gemini 2.5 Flash (Reasoning)	97	97	97	97	97	97	97	97.3%
GPT-OSS 120B	97	97	97	97	97	97	97	97.3%
Inception Mercury 2	97	97	97	97	97	97	97	97.3%
GPT-5.4 Nano (Reasoning, Low)	100	97	97	97	97	97	95	97.3%
DeepSeek-V2 Chat	100	100	100	95	95	95	95	96.9%
Grok 4.20 (Beta)	100	100	100	95	95	95	95	96.9%
Mistral Large	100	100	100	95	95	95	95	96.9%
Claude Opus 4.7	100	97	97	97	95	95	95	96.5%
DeepSeek V3 (2025-03-24)	100	100	100	95	95	95	92	96.5%
Qwen 3.5 35B	100	100	97	97	95	92	92	96.1%
DeepSeek V3 (2024-12-26)	100	100	95	95	95	95	95	96.1%
Hermes 3 70B	100	100	95	95	95	95	95	96.1%
GPT-5 Nano	100	100	97	97	92	92	92	95.8%
GPT-5.4 Nano (Reasoning)	100	97	97	95	95	95	92	95.8%
ByteDance Seed 1.6 Flash	97	97	97	97	97	95	89	95.8%
Qwen 3.5 Flash	100	100	97	95	92	92	92	95.4%
GPT-5.4 Nano	97	97	97	97	95	92	92	95.4%
WizardLM 2 8x22b	100	95	95	95	95	95	95	95.4%
Ministral 3 14B	97	97	95	95	95	95	95	95.4%
Gemini 2.5 Flash Lite (Reasoning)	100	95	95	95	95	92	92	94.6%
Mistral Large 3	95	95	95	95	95	95	95	94.6%
GPT-5.4 Mini	95	95	95	95	95	95	95	94.6%
Mistral Large 2	95	95	95	95	95	95	95	94.6%
DeepSeek V3.2	95	95	95	95	95	95	95	94.6%
DeepSeek V4 Flash	95	95	95	95	95	95	95	94.6%
GPT-4o Mini (temp=1)	95	95	95	95	95	95	95	94.6%
Mistral Small 3.2 24B	95	95	95	95	95	95	95	94.6%
Gemma 3 12B	95	95	95	95	95	95	95	94.6%
GPT-4o Mini (temp=0)	95	95	95	95	95	95	95	94.6%
Mistral Small 4	95	95	95	95	95	95	95	94.6%
Ministral 3 8B	95	95	95	95	95	95	95	94.6%
Mistral NeMO	95	95	95	95	95	95	95	94.6%
Grok 4.3	95	95	95	95	95	95	92	94.2%
Ministral 8B	95	95	95	95	95	95	92	94.2%
Z.AI GLM 4.7 Flash	100	97	95	95	95	86	86	93.4%
ByteDance Seed 2.0 Mini	95	95	95	95	95	92	86	93.1%
Nemotron 3 Super	97	97	95	95	92	92	78	92.3%
Cohere Command R+ (Aug. 2024)	95	95	95	95	95	95	78	92.3%
Llama 3.1 8B	95	95	95	95	95	92	81	92.3%
Mistral Small 4 (Reasoning)	100	100	97	97	95	92	49	90.0%
LFM2 24B	86	86	86	86	86	86	86	86.5%
Gemini 2.5 Flash Lite	86	86	86	86	86	86	51	81.5%
DeepSeek V3.1	95	95	95	95	95	95	0	81.1%
Nemotron 3 Nano	97	97	95	95	95	41	30	78.4%
Xiaomi MIMO v2.5 Pro	100	100	100	100	100	0	0	71.4%
Rocinante 12B	100	92	89	86	51	51	0	67.2%
Arcee AI: Trinity Large (Preview)	100	0	0	0	0	0	0	14.3%
GPT-4.1 Nano	51	3	3	3	3	3	0	9.3%

▼

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

Model	# 1	# 2	# 3	# 4	# 5	# 6	# 7	Avg ▼
Qwen3.6 Max Preview	100	100	100	100	100	100	100	100.0%
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5.1	100	100	100	100	100	100	100	100.0%
Z.AI GLM 5 Turbo	100	100	100	100	100	100	100	100.0%

Name replacement accuracy

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

Generic Prompt

Character rename: Elena->Mirabel, Gregor->Aldric

Combined: 3rd person past → 1st person present

Expand all contractions

Location rename: market square, outer ring, bridge, northern mines

Multi-character gender swap: Priya(F)->Rohan(M), Mara unchanged

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