Adverbs in dialogue tags

Test: Bad Writing Habits

Avg. Score

89.8%

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	Stealth: Aurora Alpha	99.3%	$0.0000	9.8s	89%
2	ByteDance Seed 1.6 Flash	99.2%	$0.0013	27.3s	88%
3	Grok 4.1 Fast	98.7%	$0.0018	37.8s	89%
4	Ministral 8B	98.2%	$0.0004	10.4s	78%
5	Qwen 2.5 72B	98.6%	$0.0010	36.7s	81%
6	Gemini 3 Flash (Preview)	98.5%	$0.0078	19.6s	78%
7	o4 Mini	97.4%	$0.015	25.7s	83%
8	Ministral 3 3B	96.8%	$0.0005	11.1s	70%
9	Mistral NeMO	96.1%	$0.0005	10.1s	69%
10	Qwen 3.5 Plus (2026-02-15)	95.9%	$0.0060	31.5s	70%
11	Z.AI GLM 4.6	96.9%	$0.0065	51.5s	73%
12	Ministral 3B	94.8%	$0.0001	8.1s	63%
13	GPT-4o, May 13th (temp=0)	96.0%	$0.035	14.1s	73%
14	WizardLM 2 8x22b	96.0%	$0.0026	1.8m	80%
15	Mistral Small Creative	94.2%	$0.0007	9.1s	57%
16	Gemini 2.5 Pro	96.2%	$0.036	36.2s	73%
17	GPT-4o Mini (temp=0)	94.2%	$0.0012	34.8s	62%
18	Writer: Palmyra X5	94.2%	$0.011	22.0s	62%
19	Gemini 2.5 Flash Lite	91.0%	$0.0009	9.5s	57%
20	Z.AI GLM 4.5	91.8%	$0.0051	42.1s	65%
21	GPT-4o, Aug. 6th (temp=0)	93.1%	$0.023	22.7s	63%
22	Mistral Medium 3.1	93.4%	$0.0048	36.5s	58%
23	Grok 4 Fast	92.2%	$0.0017	24.1s	55%
24	Gemini 2.5 Flash	90.8%	$0.0052	10.6s	53%
25	GPT-4.1	94.5%	$0.018	44.7s	60%
26	Cohere Command R+ (Aug. 2024)	94.0%	$0.020	52.5s	63%
27	Ministral 3 8B	90.8%	$0.0008	19.6s	52%
28	Z.AI GLM 4.7	94.2%	$0.010	1.4m	64%
29	Gemini 3 Pro (Preview)	95.7%	$0.055	54.4s	71%
30	Mistral Large 2	92.0%	$0.013	29.4s	54%
31	Qwen 3.5 397B A17B	98.9%	$0.014	3.0m	79%
32	o4 Mini High	93.3%	$0.025	47.2s	59%
33	Claude Sonnet 4.5	93.5%	$0.035	38.1s	60%
34	Gemma 3 4B	89.6%	$0.0002	20.0s	49%
35	DeepSeek V3.2	94.2%	$0.0014	1.9m	64%
36	Mistral Large	92.1%	$0.014	30.9s	50%
37	GPT-5 Mini	90.7%	$0.0100	57.4s	57%
38	GPT-5.1	95.4%	$0.054	1.8m	78%
39	GPT-5.2	96.7%	$0.056	1.5m	73%
40	Z.AI GLM 4.7 Flash	91.8%	$0.0017	1.2m	55%
41	Ministral 3 14B	88.8%	$0.0007	11.7s	43%
42	Z.AI GLM 5	90.6%	$0.0084	1.2m	57%
43	Llama 3.1 8B	90.9%	$0.0003	1.3m	51%
44	Claude Sonnet 4.6	91.5%	$0.031	39.3s	52%
45	Gemma 3 27B	87.8%	$0.0006	52.6s	49%
46	Mistral Large 3	88.9%	$0.0033	30.3s	41%
47	Claude 3.5 Haiku	87.3%	$0.0035	10.8s	37%
48	Claude 3 Haiku	85.5%	$0.0025	14.9s	39%
49	GPT-4o Mini (temp=1)	82.8%	$0.0012	34.8s	47%
50	Arcee AI: Trinity Mini	83.6%	$0.0003	9.2s	39%
51	DeepSeek V3 (2025-03-24)	86.7%	$0.0014	39.4s	41%
52	ByteDance Seed 1.6	94.4%	$0.013	2.5m	59%
53	Claude Opus 4.6	95.2%	$0.078	1.2m	63%
54	Llama 3.1 70B	85.1%	$0.0015	29.4s	38%
55	GPT-5	96.8%	$0.065	2.8m	76%
56	Claude Opus 4.5	91.5%	$0.070	53.4s	58%
57	DeepSeek V3.1	88.1%	$0.0020	1.8m	48%
58	Llama 3.1 Nemotron 70B	84.1%	$0.0038	31.7s	36%
59	Minimax M2.5	84.3%	$0.0034	1.3m	45%
60	DeepSeek V3 (2024-12-26)	84.5%	$0.0021	54.6s	39%
61	DeepSeek-V2 Chat	84.0%	$0.0021	53.3s	39%
62	Gemma 3 12B	82.5%	$0.0004	41.3s	36%
63	Grok 4	92.2%	$0.048	1.7m	53%
64	MoonshotAI: Kimi K2.5	93.6%	$0.019	3.2m	60%
65	Claude Sonnet 4	84.8%	$0.032	43.7s	41%
66	Claude 3.5 Sonnet	86.1%	$0.048	35.5s	42%
67	GPT-4o, May 13th (temp=1)	83.0%	$0.033	14.4s	37%
68	Rocinante 12B	81.7%	$0.0014	38.4s	32%
69	GPT-4o, Aug. 6th (temp=1)	80.5%	$0.018	24.4s	37%
70	Claude 3.7 Sonnet	82.8%	$0.042	46.7s	46%
71	Claude Haiku 4.5	76.9%	$0.011	21.6s	32%
72	Arcee AI: Trinity Large (Preview)	78.8%	$0.0000	43.6s	28%
73	Gemini 3.1 Pro (Preview)	93.3%	$0.107	1.8m	59%
74	GPT-4.1 Mini	72.0%	$0.0027	19.0s	29%
75	Hermes 3 405B	74.1%	$0.0032	53.2s	22%
76	GPT-5 Nano	72.2%	$0.0042	1.4m	27%
77	Mistral Small 3.2 24B	92.8%	$0.0069	5.7m	50%
78	Hermes 3 70B	67.6%	$0.0010	1.2m	19%
79	Claude Opus 4	91.7%	$0.209	1.4m	60%
80	GPT-4.1 Nano	50.7%	$0.0007	13.3s	8%
89.76%

Individual Scenarios

Detailed Writing Rules

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	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%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
DeepSeek V3.1	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 27B	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	92	98.4%
Claude Sonnet 4.6	100	100	100	100	91	98.2%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	89	97.8%
o4 Mini	100	100	100	100	86	97.1%
Mistral Large	100	100	100	100	86	97.1%
Mistral Small 3.2 24B	100	100	100	100	85	96.9%
WizardLM 2 8x22b	100	100	100	92	84	95.2%
Llama 3.1 70B	100	100	100	100	75	95.0%
Gemini 2.5 Flash Lite	100	100	100	100	75	95.0%
Gemma 3 12B	100	100	100	100	70	93.9%
Arcee AI: Trinity Large (Preview)	100	100	100	100	70	93.9%
DeepSeek V3.2	100	100	100	100	67	93.3%
Mistral Large 3	100	100	100	100	65	93.0%
Claude Opus 4	100	100	100	86	75	92.3%
GPT-4.1 Mini	100	100	100	100	57	91.4%
Ministral 3B	100	100	100	100	57	91.4%
Gemini 3.1 Pro (Preview)	100	100	100	100	52	90.4%
Claude Haiku 4.5	100	100	100	100	42	88.4%
Llama 3.1 Nemotron 70B	100	100	100	100	33	86.7%
Claude 3.7 Sonnet	100	100	100	70	64	86.6%
GPT-5 Nano	100	100	100	87	44	86.1%
GPT-4o Mini (temp=1)	100	100	100	85	40	84.9%
GPT-5 Mini	100	100	100	100	24	84.7%
GPT-4o, May 13th (temp=1)	100	100	100	82	40	84.5%
DeepSeek V3 (2024-12-26)	100	100	100	67	46	82.6%
Claude Sonnet 4	100	100	100	57	52	81.8%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	0	80.0%
Ministral 3 3B	100	100	100	100	0	80.0%
DeepSeek V3 (2025-03-24)	100	100	100	89	0	77.8%
Minimax M2.5	100	97	87	57	42	76.7%
Cohere Command R+ (Aug. 2024)	100	100	100	57	18	75.1%
ByteDance Seed 1.6	100	100	100	75	0	75.0%
Z.AI GLM 4.5	100	100	100	42	29	74.1%
Z.AI GLM 5	100	100	100	57	10	73.3%
MoonshotAI: Kimi K2.5	100	100	100	18	0	63.6%
Hermes 3 405B	100	100	89	0	0	57.8%
Hermes 3 70B	100	100	33	0	0	46.7%
GPT-4.1 Nano	100	100	33	0	0	46.7%
Rocinante 12B	100	52	0	0	0	30.4%

▼

Horror: alone in an eerie place at night

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
GPT-4.1 Nano	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	95	98.9%
Gemini 3.1 Pro (Preview)	100	100	100	100	89	97.8%
Claude Opus 4.5	100	100	100	100	89	97.8%
ByteDance Seed 1.6 Flash	100	100	100	100	89	97.8%
Z.AI GLM 4.5	100	100	100	100	82	96.5%
GPT-5.1	100	100	100	100	75	95.0%
Gemma 3 12B	100	100	100	100	57	91.4%
Cohere Command R+ (Aug. 2024)	100	100	100	100	57	91.4%
Llama 3.1 8B	100	100	100	75	75	90.0%
Grok 4	100	100	100	100	33	86.7%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	33	86.7%
Claude 3.7 Sonnet	100	100	89	82	50	84.2%
ByteDance Seed 1.6	100	100	100	100	18	83.6%
GPT-4.1	100	100	100	100	0	80.0%
Grok 4 Fast	100	100	100	100	0	80.0%
DeepSeek-V2 Chat	100	100	100	100	0	80.0%
Claude Haiku 4.5	100	100	100	100	0	80.0%
DeepSeek V3.1	100	100	100	100	0	80.0%
Mistral Large 2	100	100	100	100	0	80.0%
Llama 3.1 Nemotron 70B	100	100	100	100	0	80.0%
Hermes 3 70B	100	100	100	100	0	80.0%
Rocinante 12B	100	100	100	100	0	80.0%
Ministral 3 8B	100	100	100	89	0	77.8%
Writer: Palmyra X5	100	100	100	82	0	76.5%
Mistral Large	100	100	100	82	0	76.5%
Hermes 3 405B	100	100	100	75	0	75.0%
Gemma 3 4B	100	100	89	75	0	72.8%
Gemini 2.5 Flash	100	100	100	57	0	71.4%
GPT-5 Nano	100	100	100	26	0	65.2%
GPT-4.1 Mini	100	100	67	57	0	64.8%
GPT-5 Mini	100	100	100	0	0	60.0%
Claude Opus 4.6	100	100	100	0	0	60.0%
Minimax M2.5	100	100	100	0	0	60.0%
DeepSeek V3 (2024-12-26)	100	100	100	0	0	60.0%
Ministral 3 14B	100	100	100	0	0	60.0%
Claude 3 Haiku	100	100	100	0	0	60.0%
Claude Sonnet 4	100	100	67	0	0	53.3%
GPT-4o, May 13th (temp=1)	100	100	33	18	0	50.3%
GPT-4o Mini (temp=1)	67	33	33	33	0	33.3%

▼

Literary fiction: old friends reunite

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
GPT-5 Nano	100	100	100	100	97	99.3%
Grok 4 Fast	100	100	100	100	97	99.3%
DeepSeek V3 (2024-12-26)	100	100	100	100	95	98.9%
Mistral Small Creative	100	100	100	100	93	98.6%
Gemini 2.5 Flash	100	100	100	100	92	98.4%
Gemini 3 Pro (Preview)	100	100	100	100	90	98.1%
Mistral NeMO	100	100	100	100	89	97.8%
Z.AI GLM 5	100	100	100	100	87	97.5%
Grok 4.1 Fast	100	100	100	100	85	96.9%
Claude Sonnet 4.6	100	100	100	100	82	96.5%
GPT-4o, May 13th (temp=0)	100	100	100	100	82	96.5%
Gemini 2.5 Pro	100	100	100	100	80	96.0%
Llama 3.1 70B	100	100	100	89	89	95.6%
Claude Sonnet 4	100	100	100	100	75	95.0%
Llama 3.1 8B	100	100	100	100	75	95.0%
Z.AI GLM 4.6	100	100	97	97	80	95.0%
DeepSeek-V2 Chat	100	100	100	86	86	94.3%
Claude Opus 4.5	100	100	100	100	71	94.2%
Claude Opus 4	100	100	100	100	65	93.0%
Grok 4	100	100	100	100	62	92.4%
GPT-4o Mini (temp=0)	100	100	92	89	79	91.9%
WizardLM 2 8x22b	100	100	100	92	64	91.3%
Mistral Medium 3.1	100	100	100	100	43	88.6%
Arcee AI: Trinity Mini	100	100	100	100	42	88.4%
Claude 3.7 Sonnet	100	100	90	82	69	88.2%
Mistral Large 3	100	100	100	100	38	87.6%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	38	87.6%
DeepSeek V3.2	100	100	100	100	33	86.7%
Gemini 2.5 Flash Lite	100	100	100	75	57	86.4%
GPT-4.1 Mini	100	100	97	71	62	85.9%
Gemma 3 12B	100	100	100	82	46	85.7%
GPT-4o, Aug. 6th (temp=1)	100	100	100	67	57	84.8%
o4 Mini High	100	100	100	100	21	84.3%
GPT-4o, May 13th (temp=1)	100	100	93	63	60	83.1%
Ministral 3 8B	100	100	100	65	43	81.6%
DeepSeek V3.1	100	100	100	79	29	81.5%
Rocinante 12B	100	100	100	100	0	80.0%
Ministral 3 14B	100	100	100	67	31	79.4%
Z.AI GLM 4.5	100	100	77	53	51	76.1%
Gemma 3 4B	100	100	100	80	0	76.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	39	36	75.0%
Claude 3 Haiku	100	100	100	36	33	73.9%
Hermes 3 405B	100	100	100	67	0	73.3%
Gemma 3 27B	100	95	95	48	18	71.2%
GPT-4o Mini (temp=1)	100	93	93	57	0	68.7%
Hermes 3 70B	100	100	89	46	0	67.0%
Minimax M2.5	100	89	73	39	0	60.2%
Arcee AI: Trinity Large (Preview)	100	100	50	33	0	56.7%
Llama 3.1 Nemotron 70B	100	100	75	0	0	55.0%
Claude Haiku 4.5	100	71	40	3	0	42.8%
GPT-4.1 Nano	40	24	0	0	0	12.7%

▼

Mystery: examining a crime scene

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Claude Haiku 4.5	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
DeepSeek V3.1	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
Gemma 3 27B	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	89	97.8%
Ministral 3 3B	100	100	100	100	82	96.5%
Gemini 2.5 Flash	100	100	100	100	80	96.0%
Mistral NeMO	100	100	100	100	80	96.0%
Claude 3.7 Sonnet	100	100	100	100	79	95.8%
Writer: Palmyra X5	100	100	100	100	79	95.8%
Claude Sonnet 4.5	100	100	100	100	78	95.5%
Ministral 3 8B	100	100	100	100	75	95.0%
WizardLM 2 8x22b	100	100	100	97	77	94.6%
GPT-4o, May 13th (temp=1)	100	100	100	100	70	93.9%
Claude Sonnet 4.6	100	100	100	100	67	93.3%
GPT-4o Mini (temp=0)	100	100	100	100	64	92.9%
GPT-5 Nano	100	100	100	100	62	92.4%
Minimax M2.5	100	100	100	85	73	91.5%
GPT-4o Mini (temp=1)	100	100	100	100	57	91.4%
Cohere Command R+ (Aug. 2024)	100	100	100	100	50	90.0%
Grok 4 Fast	100	100	100	79	71	90.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	31	86.1%
Mistral Medium 3.1	100	100	100	78	52	85.9%
Z.AI GLM 4.5	100	100	100	100	29	85.7%
Arcee AI: Trinity Large (Preview)	100	100	100	100	25	84.9%
Claude 3.5 Sonnet	100	100	100	86	33	83.8%
Mistral Large	100	100	100	100	18	83.6%
Hermes 3 70B	100	100	100	100	18	83.6%
GPT-4o, Aug. 6th (temp=1)	100	100	100	89	29	83.5%
Gemma 3 12B	100	100	100	100	0	80.0%
Mistral Small 3.2 24B	100	100	100	100	0	80.0%
Claude 3 Haiku	100	100	100	100	0	80.0%
DeepSeek V3 (2025-03-24)	100	100	100	95	0	78.9%
Gemini 2.5 Flash Lite	100	100	100	40	0	68.0%
Hermes 3 405B	100	100	71	57	0	65.6%
GPT-4.1 Mini	100	79	72	57	14	64.4%
Claude 3.5 Haiku	100	100	100	0	0	60.0%
GPT-4.1 Nano	71	67	0	0	0	27.5%

▼

Romance: separated couple reunites

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	97	99.5%
Mistral Small 3.2 24B	100	100	100	100	97	99.5%
Grok 4	100	100	100	100	95	98.9%
DeepSeek V3 (2025-03-24)	100	100	100	100	95	98.9%
Grok 4 Fast	100	100	100	100	95	98.9%
Ministral 3 14B	100	100	100	100	89	97.8%
WizardLM 2 8x22b	100	100	98	95	95	97.5%
Gemini 2.5 Flash	100	100	100	100	87	97.4%
Claude Sonnet 4.5	100	100	100	100	86	97.3%
Z.AI GLM 4.7 Flash	100	100	100	100	79	95.8%
o4 Mini High	100	100	100	100	78	95.5%
DeepSeek V3 (2024-12-26)	100	100	100	98	79	95.4%
DeepSeek V3.1	100	100	100	100	75	95.0%
Llama 3.1 8B	100	100	100	100	75	95.0%
Mistral Large 2	100	100	100	86	82	93.6%
GPT-5 Mini	100	100	100	98	69	93.4%
Minimax M2.5	100	100	100	100	65	93.0%
DeepSeek V3.2	100	100	100	100	60	92.1%
Claude Haiku 4.5	100	100	97	84	78	91.8%
Z.AI GLM 4.5	100	100	100	95	63	91.6%
Claude Sonnet 4	100	100	100	100	52	90.4%
Mistral Medium 3.1	100	100	100	97	55	90.2%
Gemma 3 12B	100	100	100	100	43	88.6%
Ministral 3 8B	100	100	100	100	26	85.2%
GPT-4o, May 13th (temp=0)	100	100	97	85	24	81.1%
GPT-4.1 Mini	100	100	97	55	50	80.4%
GPT-4o, Aug. 6th (temp=1)	100	100	100	89	13	80.3%
Claude 3.5 Haiku	100	100	100	100	0	80.0%
Arcee AI: Trinity Mini	100	100	100	100	0	80.0%
GPT-4o Mini (temp=1)	100	95	89	85	28	79.2%
Gemma 3 4B	100	100	97	57	33	77.6%
Rocinante 12B	100	100	100	82	0	76.5%
GPT-5 Nano	100	100	93	68	15	75.2%
GPT-4o, Aug. 6th (temp=0)	100	100	67	57	52	75.1%
Claude 3 Haiku	100	100	100	54	18	74.4%
Claude 3.7 Sonnet	100	100	100	43	26	73.8%
Hermes 3 70B	100	100	100	13	0	62.5%
Gemma 3 27B	100	100	57	46	5	61.6%
DeepSeek-V2 Chat	100	79	62	8	0	49.8%
Llama 3.1 70B	100	75	67	0	0	48.3%
Llama 3.1 Nemotron 70B	100	75	33	18	0	45.3%
Arcee AI: Trinity Large (Preview)	100	100	0	0	0	40.0%
GPT-4.1 Nano	62	18	13	0	0	18.6%

▼

Thriller: chase through city streets

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
Mistral Large	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	97	99.5%
DeepSeek V3 (2024-12-26)	100	100	100	100	95	98.9%
Gemma 3 27B	100	100	100	100	95	98.9%
WizardLM 2 8x22b	100	100	100	100	95	98.9%
Z.AI GLM 5	100	100	100	100	92	98.4%
o4 Mini	100	100	100	100	89	97.8%
Claude Opus 4	100	100	100	100	89	97.8%
Mistral Medium 3.1	100	100	100	100	82	96.5%
Gemini 3 Flash (Preview)	100	100	100	100	75	95.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	71	94.2%
Minimax M2.5	100	100	100	92	75	93.4%
o4 Mini High	100	100	100	100	67	93.3%
Claude Sonnet 4.5	100	100	100	100	62	92.4%
Grok 4 Fast	100	100	100	100	57	91.4%
MoonshotAI: Kimi K2.5	100	100	100	100	52	90.4%
Z.AI GLM 4.6	100	100	100	100	33	86.7%
Llama 3.1 70B	100	100	100	100	33	86.7%
Claude Sonnet 4	100	100	100	92	33	85.0%
Hermes 3 70B	100	100	100	100	18	83.6%
Gemini 2.5 Flash	100	100	100	100	10	81.9%
DeepSeek V3.1	100	100	100	57	46	80.7%
DeepSeek-V2 Chat	100	100	100	89	14	80.6%
Grok 4	100	100	100	100	0	80.0%
GPT-4.1	100	100	100	100	0	80.0%
GPT-5 Nano	100	100	100	100	0	80.0%
Claude Haiku 4.5	100	100	100	100	0	80.0%
DeepSeek V3.2	100	100	100	100	0	80.0%
Hermes 3 405B	100	100	100	100	0	80.0%
Arcee AI: Trinity Mini	100	100	100	100	0	80.0%
Gemma 3 4B	100	100	100	100	0	80.0%
Ministral 8B	100	100	100	100	0	80.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	79	0	75.8%
GPT-4.1 Mini	100	100	100	57	18	75.1%
Gemma 3 12B	100	100	100	52	18	74.0%
GPT-4.1 Nano	100	100	100	33	0	66.7%
Rocinante 12B	100	100	100	33	0	66.7%
Claude 3.7 Sonnet	100	100	100	8	0	61.5%
GPT-4o, May 13th (temp=1)	100	100	100	0	0	60.0%
Claude 3.5 Sonnet	100	100	82	0	0	56.5%
GPT-4o, Aug. 6th (temp=1)	100	100	18	0	0	43.6%

genre

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
DeepSeek-V2 Chat	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	97	99.5%
Claude 3 Haiku	100	100	100	100	97	99.5%
Ministral 8B	100	100	100	100	97	99.5%
ByteDance Seed 1.6 Flash	100	100	100	100	97	99.3%
o4 Mini High	100	100	100	100	94	98.9%
GPT-4.1	100	100	100	100	94	98.9%
Claude Sonnet 4	100	100	100	100	91	98.2%
Grok 4	100	100	100	100	89	97.8%
Grok 4.1 Fast	100	100	100	100	89	97.8%
Gemini 2.5 Pro	100	100	100	100	88	97.5%
GPT-4o, May 13th (temp=1)	100	100	100	100	88	97.5%
WizardLM 2 8x22b	100	100	100	100	88	97.5%
Grok 4 Fast	100	100	100	100	86	97.1%
Claude 3.7 Sonnet	100	100	100	100	83	96.7%
Mistral Large 3	100	100	100	100	82	96.5%
DeepSeek V3.2	100	100	100	94	88	96.4%
DeepSeek V3.1	100	100	100	100	79	95.7%
Z.AI GLM 5	100	100	100	100	75	95.0%
ByteDance Seed 1.6	100	100	100	100	75	95.0%
GPT-4.1 Mini	100	100	100	91	79	93.9%
Gemini 3 Pro (Preview)	100	100	100	94	75	93.9%
Claude Opus 4.5	100	100	100	100	67	93.3%
Writer: Palmyra X5	100	100	100	100	67	93.3%
Claude Haiku 4.5	100	100	94	91	79	92.8%
DeepSeek V3 (2025-03-24)	100	100	100	97	67	92.8%
Claude Opus 4	100	100	100	92	70	92.5%
Ministral 3 14B	100	100	100	100	57	91.4%
Claude Opus 4.6	100	100	89	85	82	91.1%
DeepSeek V3 (2024-12-26)	100	100	100	85	67	90.3%
Rocinante 12B	100	100	100	100	50	90.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	47	89.3%
GPT-4o, Aug. 6th (temp=1)	100	100	100	79	67	89.0%
Z.AI GLM 4.5	100	100	97	80	63	88.1%
Llama 3.1 Nemotron 70B	100	100	100	100	39	87.8%
Minimax M2.5	100	97	91	83	67	87.7%
Mistral Medium 3.1	100	100	100	100	38	87.6%
Arcee AI: Trinity Mini	100	100	100	79	59	87.5%
GPT-5	100	100	99	90	47	87.1%
Gemma 3 12B	100	100	100	100	33	86.7%
GPT-4o Mini (temp=1)	100	100	100	71	56	85.5%
Gemini 2.5 Flash	100	100	100	100	25	85.0%
Llama 3.1 70B	100	100	100	50	50	80.0%
GPT-5.1	100	100	83	78	38	79.8%
Hermes 3 405B	100	100	89	62	45	79.2%
Z.AI GLM 4.7	100	97	82	79	36	78.8%
Claude Sonnet 4.6	100	100	100	79	7	77.1%
Arcee AI: Trinity Large (Preview)	100	100	90	54	39	76.5%
GPT-4.1 Nano	88	88	88	59	59	76.1%
GPT-5 Nano	100	94	71	69	41	75.2%
Gemini 3.1 Pro (Preview)	100	100	100	67	0	73.3%
Hermes 3 70B	100	100	94	39	32	73.1%
Claude 3.5 Haiku	100	100	100	7	0	61.4%

▼

Horror: alone in an eerie place at night

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=1)	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Hermes 3 405B	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Gemma 3 27B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	94	98.9%
Claude Opus 4	100	100	100	100	88	97.5%
GPT-4o Mini (temp=1)	100	100	100	88	79	93.2%
GPT-5 Mini	100	100	100	88	73	92.1%
o4 Mini	100	100	100	100	50	90.0%
Qwen 2.5 72B	100	100	100	100	50	90.0%
ByteDance Seed 1.6 Flash	100	100	100	100	46	89.2%
Stealth: Aurora Alpha	100	100	100	88	50	87.5%
Arcee AI: Trinity Large (Preview)	100	100	100	88	50	87.5%
Hermes 3 70B	100	100	100	67	67	86.7%
GPT-5	100	100	100	67	55	84.3%
Z.AI GLM 5	100	100	100	94	25	83.9%
GPT-5.1	100	100	83	82	50	83.0%
Claude 3.7 Sonnet	100	100	91	88	25	80.7%
Claude Opus 4.6	100	100	100	100	0	80.0%
Claude Opus 4.5	100	100	100	100	0	80.0%
Gemini 2.5 Pro	100	100	100	100	0	80.0%
Z.AI GLM 4.7	100	100	100	100	0	80.0%
Claude Sonnet 4.6	100	100	100	100	0	80.0%
Gemini 3 Flash (Preview)	100	100	100	100	0	80.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	0	80.0%
Grok 4 Fast	100	100	100	100	0	80.0%
Claude 3.5 Sonnet	100	100	100	100	0	80.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	0	80.0%
DeepSeek V3.1	100	100	100	100	0	80.0%
GPT-4o Mini (temp=0)	100	100	100	100	0	80.0%
Gemma 3 12B	100	100	100	100	0	80.0%
Llama 3.1 Nemotron 70B	100	100	100	100	0	80.0%
Mistral Small Creative	100	100	100	100	0	80.0%
GPT-4.1 Nano	100	100	100	100	0	80.0%
GPT-5.2	100	100	100	88	7	78.9%
Claude Sonnet 4	100	100	100	94	0	78.9%
Writer: Palmyra X5	100	100	100	79	0	75.7%
o4 Mini High	100	100	79	67	0	69.0%
Llama 3.1 8B	100	100	100	39	0	67.8%
GPT-5 Nano	100	100	100	25	0	65.0%
DeepSeek-V2 Chat	100	100	100	25	0	65.0%
Claude Haiku 4.5	100	100	100	25	0	65.0%
GPT-4.1 Mini	100	100	100	25	0	65.0%
Minimax M2.5	100	100	100	0	0	60.0%
Claude Sonnet 4.5	100	100	100	0	0	60.0%
Mistral Large	100	100	100	0	0	60.0%
Mistral Medium 3.1	100	100	95	0	0	58.9%
Z.AI GLM 4.7 Flash	100	100	79	0	0	55.7%
Mistral Large 2	100	100	25	0	0	45.0%
Mistral Large 3	100	100	0	0	0	40.0%
GPT-4o, Aug. 6th (temp=1)	100	100	0	0	0	40.0%
Ministral 3 14B	100	100	0	0	0	40.0%

▼

Literary fiction: old friends reunite

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	95	99.1%
Writer: Palmyra X5	100	100	100	100	95	99.1%
DeepSeek V3 (2025-03-24)	100	100	100	100	94	98.9%
Llama 3.1 Nemotron 70B	100	100	100	100	94	98.9%
GPT-4.1	100	100	100	100	93	98.7%
Claude Opus 4.6	100	100	100	96	94	98.2%
Gemini 3 Flash (Preview)	100	100	100	97	92	97.9%
GPT-5	100	100	100	100	87	97.3%
GPT-4o Mini (temp=0)	100	100	100	100	86	97.2%
Z.AI GLM 4.6	100	100	100	100	85	97.0%
Gemini 2.5 Flash Lite	100	100	100	100	85	97.0%
Claude Sonnet 4.6	100	100	100	100	84	96.9%
Ministral 8B	100	100	100	100	79	95.7%
Claude Sonnet 4.5	100	100	93	92	91	95.4%
Claude 3 Haiku	100	100	97	91	88	95.2%
Gemini 2.5 Flash	100	100	100	100	71	94.2%
o4 Mini High	100	100	100	90	81	94.2%
Mistral Medium 3.1	100	100	100	89	80	93.8%
Qwen 3.5 Plus (2026-02-15)	100	100	100	89	77	93.2%
Z.AI GLM 4.5	100	100	100	93	72	93.1%
Z.AI GLM 4.7	100	100	100	94	68	92.5%
Ministral 3 8B	100	100	100	100	62	92.4%
Z.AI GLM 4.7 Flash	100	100	97	85	79	92.3%
WizardLM 2 8x22b	100	100	100	88	73	92.1%
Llama 3.1 8B	100	100	100	100	59	91.8%
GPT-5.1	100	100	99	93	64	91.2%
Gemini 2.5 Pro	100	100	100	85	65	90.1%
GPT-5.2	100	100	100	86	63	89.8%
o4 Mini	100	100	100	81	64	89.0%
Minimax M2.5	100	100	100	97	42	87.9%
Grok 4.1 Fast	100	97	89	85	57	85.6%
GPT-4o, Aug. 6th (temp=0)	100	94	79	77	75	85.1%
Qwen 2.5 72B	100	100	100	100	25	85.0%
DeepSeek V3.2	100	100	90	75	59	84.9%
Mistral Large 2	100	100	100	73	50	84.6%
MoonshotAI: Kimi K2.5	100	100	100	91	31	84.4%
Mistral Large 3	100	100	89	78	50	83.3%
Ministral 3B	100	100	100	100	15	83.0%
Claude Sonnet 4	100	100	85	70	57	82.6%
Z.AI GLM 5	100	100	88	67	57	82.2%
Claude Opus 4	96	91	88	69	67	82.1%
Arcee AI: Trinity Large (Preview)	100	100	85	75	45	80.9%
DeepSeek V3 (2024-12-26)	100	100	100	88	14	80.4%
Qwen 3.5 397B A17B	100	100	100	100	0	80.0%
Mistral Small 3.2 24B	100	100	100	100	0	80.0%
Mistral Small Creative	100	100	86	80	34	80.0%
Claude 3.5 Sonnet	100	92	83	73	50	79.7%
GPT-5 Nano	100	96	83	69	42	78.2%
Ministral 3 3B	100	100	100	52	33	77.0%
GPT-5 Mini	100	100	100	44	39	76.6%
GPT-4o Mini (temp=1)	100	94	79	62	41	75.3%
Hermes 3 70B	100	100	97	64	12	74.7%
GPT-4o, Aug. 6th (temp=1)	94	83	80	59	55	74.4%
Rocinante 12B	100	100	100	62	0	72.4%
Ministral 3 14B	100	90	78	55	37	72.0%
Gemini 3 Pro (Preview)	100	79	67	61	47	70.5%
Claude Opus 4.5	100	81	80	46	44	70.0%
Claude Haiku 4.5	100	100	75	57	0	66.4%
Hermes 3 405B	100	100	97	25	0	64.5%
GPT-4o, May 13th (temp=1)	100	100	62	53	0	63.0%
Grok 4 Fast	100	85	70	54	0	61.6%
DeepSeek V3.1	89	71	59	50	39	61.6%
Gemini 3.1 Pro (Preview)	100	100	100	0	0	60.0%
Claude 3.7 Sonnet	75	74	69	51	27	59.0%
Claude 3.5 Haiku	100	100	39	25	25	57.8%
Grok 4	100	100	82	0	0	56.5%
Arcee AI: Trinity Mini	100	100	54	25	0	55.8%
DeepSeek-V2 Chat	100	79	54	42	0	54.9%
Gemma 3 12B	100	100	46	18	0	52.9%
Gemma 3 27B	89	78	43	31	22	52.5%
Gemma 3 4B	100	64	28	0	0	38.2%
GPT-4.1 Mini	67	47	3	0	0	23.2%
GPT-4.1 Nano	53	20	0	0	0	14.6%

▼

Mystery: examining a crime scene

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
DeepSeek V3.1	100	100	100	100	100	100.0%
Ministral 3 14B	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	97	99.5%
GPT-4o Mini (temp=0)	100	100	100	100	97	99.5%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	96	99.3%
Grok 4.1 Fast	100	100	100	100	95	98.9%
GPT-5	100	100	100	100	94	98.9%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	94	98.9%
Gemini 3 Pro (Preview)	100	100	100	100	94	98.7%
Ministral 8B	100	100	100	100	91	98.2%
o4 Mini High	100	100	100	100	89	97.9%
Minimax M2.5	100	100	100	94	93	97.6%
Mistral Large	100	100	100	100	88	97.5%
Gemini 2.5 Pro	100	100	100	97	89	97.4%
Claude Sonnet 4.6	100	100	100	99	88	97.3%
Gemma 3 27B	100	100	100	100	86	97.1%
Z.AI GLM 4.7	100	100	100	97	87	96.8%
Llama 3.1 Nemotron 70B	100	100	100	100	83	96.7%
Mistral Small Creative	100	100	100	100	83	96.7%
Ministral 3B	100	100	100	100	83	96.7%
MoonshotAI: Kimi K2.5	100	100	100	97	83	96.1%
Z.AI GLM 5	100	100	100	93	85	95.8%
DeepSeek V3 (2024-12-26)	100	100	100	91	88	95.7%
Mistral Large 2	100	100	100	93	83	95.2%
DeepSeek-V2 Chat	100	100	100	100	73	94.6%
Arcee AI: Trinity Large (Preview)	100	100	100	100	73	94.6%
WizardLM 2 8x22b	100	100	100	100	67	93.3%
GPT-5 Mini	100	100	100	89	76	93.1%
Claude Sonnet 4.5	100	100	100	100	64	92.9%
GPT-5.1	100	100	100	86	77	92.6%
Gemini 2.5 Flash Lite	100	100	100	96	67	92.5%
Claude Opus 4.5	100	100	100	91	68	91.9%
Claude Opus 4	100	100	100	100	59	91.8%
GPT-4o, May 13th (temp=1)	100	100	100	85	71	91.3%
Claude Opus 4.6	100	100	100	89	64	90.7%
Z.AI GLM 4.5	100	100	93	85	73	90.1%
Hermes 3 70B	100	100	100	100	50	90.0%
Gemma 3 12B	100	100	100	100	46	89.2%
Rocinante 12B	100	100	100	79	67	89.0%
GPT-4o Mini (temp=1)	100	100	96	91	53	88.2%
Writer: Palmyra X5	100	100	100	89	50	87.9%
GPT-5 Nano	100	100	88	79	73	87.8%
Arcee AI: Trinity Mini	100	100	100	100	39	87.8%
DeepSeek V3.2	100	100	89	89	56	86.9%
Claude Sonnet 4	100	100	100	100	25	85.0%
Llama 3.1 70B	100	100	100	100	7	81.4%
DeepSeek V3 (2025-03-24)	100	100	100	100	0	80.0%
Mistral Large 3	100	100	100	100	0	80.0%
GPT-4.1	100	100	94	88	12	78.8%
Claude 3 Haiku	100	100	79	67	47	78.5%
Gemini 3.1 Pro (Preview)	100	100	92	57	30	75.8%
Hermes 3 405B	100	79	75	73	39	73.1%
Claude 3.7 Sonnet	100	95	79	64	27	73.0%
Claude 3.5 Sonnet	100	81	79	50	50	71.9%
Claude Haiku 4.5	99	83	62	43	39	65.1%
Claude 3.5 Haiku	100	100	100	25	0	65.0%
GPT-4.1 Mini	100	79	50	50	39	63.5%
GPT-4.1 Nano	91	59	0	0	0	30.1%

▼

Romance: separated couple reunites

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Claude Opus 4	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Mistral Small Creative	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	99	99.7%
Qwen 2.5 72B	100	100	100	100	97	99.5%
Z.AI GLM 4.6	100	100	100	100	91	98.2%
GPT-5.2	100	100	100	100	89	97.9%
Grok 4 Fast	100	100	100	100	89	97.8%
DeepSeek V3.2	100	100	100	100	89	97.8%
Mistral Large 2	100	100	100	100	88	97.5%
Ministral 3 14B	100	100	100	100	87	97.4%
Ministral 3B	100	100	100	100	85	97.1%
Minimax M2.5	100	100	100	95	90	97.0%
Claude 3.5 Sonnet	100	100	100	96	88	96.7%
Ministral 8B	100	100	100	99	83	96.4%
GPT-5	100	100	99	95	88	96.4%
o4 Mini High	100	100	100	100	82	96.4%
Writer: Palmyra X5	100	100	100	99	82	96.3%
Cohere Command R+ (Aug. 2024)	100	100	100	100	81	96.2%
Ministral 3 8B	100	100	100	100	80	95.9%
GPT-4o Mini (temp=0)	100	100	100	98	80	95.6%
Z.AI GLM 4.7	100	100	100	96	80	95.3%
GPT-4.1	100	100	97	91	88	95.2%
Claude Opus 4.6	100	100	100	88	88	95.0%
o4 Mini	100	100	93	91	89	94.8%
Grok 4	100	100	100	89	85	94.7%
Rocinante 12B	100	100	100	94	79	94.6%
WizardLM 2 8x22b	100	100	100	100	71	94.2%
DeepSeek V3 (2025-03-24)	100	100	97	88	85	94.1%
Gemma 3 12B	100	100	100	100	67	93.3%
Gemini 3.1 Pro (Preview)	100	100	95	89	82	93.2%
Gemini 2.5 Flash	100	100	100	94	67	92.2%
GPT-4o, Aug. 6th (temp=1)	100	100	97	94	67	91.7%
Gemini 2.5 Pro	100	100	100	100	56	91.3%
Z.AI GLM 4.5	100	100	99	99	53	90.2%
GPT-5.1	100	96	91	88	75	90.0%
Gemini 3 Pro (Preview)	100	100	100	100	50	90.0%
Gemma 3 4B	100	100	100	100	50	90.0%
Claude Opus 4.5	100	100	100	100	46	89.2%
ByteDance Seed 1.6	100	100	100	100	42	88.4%
DeepSeek V3.1	100	100	100	96	45	88.1%
Mistral NeMO	100	100	100	100	39	87.8%
GPT-4o, Aug. 6th (temp=0)	100	100	89	71	67	85.4%
Claude 3.7 Sonnet	100	100	100	73	53	85.2%
GPT-4o, May 13th (temp=0)	100	100	91	71	59	84.3%
Arcee AI: Trinity Mini	100	88	88	83	59	83.5%
Qwen 3.5 Plus (2026-02-15)	100	96	82	79	59	83.1%
Hermes 3 70B	100	100	100	70	43	82.6%
Z.AI GLM 5	100	91	87	73	57	81.6%
GPT-4o, May 13th (temp=1)	100	100	100	100	7	81.4%
Arcee AI: Trinity Large (Preview)	100	100	100	100	0	80.0%
Gemma 3 27B	100	100	91	60	49	79.9%
Claude 3 Haiku	100	100	100	99	0	79.7%
GPT-5 Mini	100	100	68	68	61	79.4%
Claude Sonnet 4.5	100	100	94	52	47	78.5%
Llama 3.1 8B	100	100	81	72	39	78.4%
Claude Sonnet 4.6	100	100	79	76	36	78.0%
Llama 3.1 Nemotron 70B	100	100	100	88	0	77.5%
DeepSeek-V2 Chat	100	86	75	71	48	75.9%
Z.AI GLM 4.7 Flash	100	94	85	57	43	75.8%
Claude Sonnet 4	100	100	79	62	31	74.4%
GPT-4o Mini (temp=1)	100	79	73	70	50	74.3%
Gemini 2.5 Flash Lite	100	92	75	59	42	73.7%
MoonshotAI: Kimi K2.5	100	100	83	35	25	68.6%
GPT-4.1 Mini	100	79	59	47	47	66.2%
Claude Haiku 4.5	97	75	65	48	20	61.0%
Mistral Small 3.2 24B	100	100	100	0	0	60.0%
GPT-5 Nano	89	74	67	29	25	56.9%
Llama 3.1 70B	100	91	59	17	0	53.4%
Hermes 3 405B	100	77	71	0	0	49.6%
DeepSeek V3 (2024-12-26)	100	100	35	0	0	46.9%
GPT-4.1 Nano	50	0	0	0	0	10.0%

▼

Thriller: chase through city streets

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Gemma 3 27B	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	97	99.5%
Cohere Command R+ (Aug. 2024)	100	100	100	100	97	99.5%
DeepSeek V3.2	100	100	100	100	94	98.9%
Claude 3.7 Sonnet	100	100	100	100	92	98.5%
Minimax M2.5	100	100	100	100	91	98.2%
ByteDance Seed 1.6	100	100	100	100	89	97.8%
o4 Mini	100	100	100	100	88	97.5%
Gemini 2.5 Pro	100	100	100	100	88	97.5%
GPT-4o, May 13th (temp=1)	100	100	100	100	88	97.5%
GPT-4o Mini (temp=1)	100	100	100	100	88	97.5%
Grok 4.1 Fast	100	100	100	100	82	96.5%
Gemma 3 12B	100	100	100	100	82	96.5%
MoonshotAI: Kimi K2.5	100	100	100	100	79	95.7%
Claude 3 Haiku	100	100	100	100	79	95.7%
Claude Sonnet 4	100	100	100	100	73	94.6%
Mistral Large 2	100	100	100	100	73	94.6%
Rocinante 12B	100	100	100	100	73	94.6%
Z.AI GLM 4.5	100	100	100	92	79	94.2%
GPT-4.1	100	100	100	88	83	94.2%
GPT-4o Mini (temp=0)	100	100	100	100	67	93.3%
Hermes 3 70B	100	100	100	100	67	93.3%
Writer: Palmyra X5	100	100	100	96	70	93.2%
Z.AI GLM 4.7	100	100	94	88	79	92.1%
GPT-4o, May 13th (temp=0)	100	100	100	91	59	90.1%
Arcee AI: Trinity Large (Preview)	100	100	100	83	67	90.0%
Claude Haiku 4.5	100	100	100	100	45	88.9%
DeepSeek V3 (2024-12-26)	100	100	100	73	67	87.9%
GPT-5.1	100	92	85	83	67	85.4%
Arcee AI: Trinity Mini	100	100	100	79	39	83.5%
GPT-4o, Aug. 6th (temp=1)	100	100	100	67	50	83.3%
GPT-5.2	100	100	100	91	25	83.2%
Claude Opus 4.5	100	100	77	73	62	82.3%
DeepSeek-V2 Chat	100	100	100	100	7	81.4%
Mistral NeMO	100	100	100	100	7	81.4%
Grok 4	100	100	100	100	0	80.0%
Mistral Large 3	100	100	100	100	0	80.0%
GPT-5	100	96	95	91	17	79.9%
Claude 3.5 Sonnet	100	100	100	79	17	79.0%
o4 Mini High	100	100	100	79	0	75.7%
DeepSeek V3.1	100	100	100	79	0	75.7%
Ministral 3 8B	100	100	100	75	0	75.0%
Hermes 3 405B	100	100	100	73	0	74.6%
GPT-4o, Aug. 6th (temp=0)	100	100	100	59	7	73.2%
GPT-5 Mini	100	100	59	50	39	69.6%
GPT-4.1 Nano	100	100	79	39	25	68.5%
GPT-4.1 Mini	100	100	100	0	0	60.0%
DeepSeek V3 (2025-03-24)	100	100	25	25	25	55.0%
Claude Opus 4	89	79	59	25	0	50.4%
GPT-5 Nano	100	50	45	0	0	38.9%

Novelcrafter Default Prompt

▼

Fantasy: entering an ancient ruin

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	100	100	100	100	100	100.0%
Claude Sonnet 4	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%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Gemma 3 27B	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	97	99.3%
Gemini 2.5 Flash	100	100	100	100	95	98.9%
Z.AI GLM 4.6	100	100	100	100	89	97.8%
Z.AI GLM 5	100	100	100	100	82	96.5%
Minimax M2.5	100	100	100	100	82	96.5%
Gemma 3 4B	100	100	100	100	82	96.5%
GPT-4o, May 13th (temp=1)	100	100	100	100	52	90.4%
DeepSeek V3.1	100	100	100	100	52	90.4%
DeepSeek V3 (2024-12-26)	100	100	100	100	46	89.2%
WizardLM 2 8x22b	100	100	100	100	46	89.2%
Rocinante 12B	100	100	100	100	42	88.4%
Arcee AI: Trinity Mini	100	100	100	89	52	88.1%
DeepSeek V3 (2025-03-24)	100	100	100	100	33	86.7%
GPT-4o, Aug. 6th (temp=1)	100	100	82	82	67	86.3%
Claude Opus 4	100	100	100	100	18	83.6%
GPT-4.1 Mini	100	100	100	100	13	82.5%
GPT-4o Mini (temp=1)	100	100	100	80	24	80.7%
ByteDance Seed 1.6	100	100	100	100	0	80.0%
Hermes 3 405B	100	100	100	95	0	78.9%
DeepSeek-V2 Chat	100	100	100	92	0	78.4%
Claude 3.7 Sonnet	100	100	97	52	40	77.7%
Llama 3.1 Nemotron 70B	100	100	100	33	33	73.3%
Claude Haiku 4.5	100	100	89	67	0	71.1%
GPT-4.1 Nano	100	89	89	0	0	55.6%
Claude 3.5 Sonnet	100	100	75	0	0	55.0%
Claude Sonnet 4.6	100	100	59	0	0	51.8%
Arcee AI: Trinity Large (Preview)	100	100	13	0	0	42.5%
GPT-5 Nano	100	62	29	0	0	38.3%
Hermes 3 70B	100	26	0	0	0	25.2%

▼

Horror: alone in an eerie place at night

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
Claude Haiku 4.5	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
DeepSeek V3.1	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	95	98.9%
GPT-4o, May 13th (temp=1)	100	100	100	100	89	97.8%
Gemma 3 4B	100	100	100	100	89	97.8%
DeepSeek V3 (2024-12-26)	100	100	100	100	75	95.0%
GPT-4.1 Mini	100	100	100	100	75	95.0%
GPT-5.2	100	100	100	100	57	91.4%
Claude Opus 4	100	100	100	100	57	91.4%
Minimax M2.5	100	100	100	100	57	91.4%
Gemma 3 27B	100	100	100	100	46	89.2%
Claude Sonnet 4.5	100	100	100	100	33	86.7%
Claude 3.7 Sonnet	100	100	100	75	57	86.4%
GPT-4o Mini (temp=1)	100	100	100	89	18	81.4%
o4 Mini High	100	100	100	100	0	80.0%
Z.AI GLM 5	100	100	100	100	0	80.0%
Z.AI GLM 4.7	100	100	100	100	0	80.0%
Gemini 3 Pro (Preview)	100	100	100	100	0	80.0%
Claude Sonnet 4	100	100	100	100	0	80.0%
Claude Sonnet 4.6	100	100	100	100	0	80.0%
GPT-4.1	100	100	100	100	0	80.0%
Z.AI GLM 4.7 Flash	100	100	100	100	0	80.0%
Grok 4 Fast	100	100	100	100	0	80.0%
Claude 3.5 Sonnet	100	100	100	100	0	80.0%
DeepSeek-V2 Chat	100	100	100	100	0	80.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	0	80.0%
Writer: Palmyra X5	100	100	100	100	0	80.0%
Gemini 2.5 Flash	100	100	100	100	0	80.0%
Hermes 3 405B	100	100	100	100	0	80.0%
Llama 3.1 70B	100	100	100	100	0	80.0%
Gemini 2.5 Flash Lite	100	100	100	100	0	80.0%
Ministral 3 8B	100	100	100	100	0	80.0%
Arcee AI: Trinity Mini	100	100	100	100	0	80.0%
GPT-4.1 Nano	100	100	100	100	0	80.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	75	0	75.0%
Grok 4	100	100	89	57	18	72.8%
Mistral Large 2	100	100	100	57	0	71.4%
Llama 3.1 Nemotron 70B	100	100	100	57	0	71.4%
Cohere Command R+ (Aug. 2024)	100	100	100	57	0	71.4%
Mistral Large	100	100	100	0	0	60.0%
Mistral Small Creative	100	100	100	0	0	60.0%
Ministral 3 14B	100	100	100	0	0	60.0%
Hermes 3 70B	100	100	46	33	0	55.9%
Mistral Large 3	100	100	0	0	0	40.0%
GPT-5 Nano	100	0	0	0	0	20.0%

▼

Literary fiction: old friends reunite

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
Claude Opus 4.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	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%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	100	100	100	100	100	100.0%
Gemma 3 27B	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	97	99.4%
DeepSeek V3.2	100	100	100	100	95	98.9%
DeepSeek-V2 Chat	100	100	100	98	95	98.6%
Grok 4 Fast	100	100	100	100	93	98.6%
Claude Sonnet 4.6	100	100	100	100	91	98.2%
GPT-5 Nano	100	100	100	100	75	95.0%
Mistral Large 2	100	100	100	100	75	95.0%
Mistral Large	100	100	100	100	64	92.9%
Gemma 3 4B	100	100	89	86	82	91.4%
Z.AI GLM 5	100	100	100	77	73	90.0%
Gemini 2.5 Pro	100	100	100	97	52	89.9%
Z.AI GLM 4.5	100	100	100	100	49	89.8%
GPT-4o Mini (temp=1)	100	100	97	93	48	87.8%
GPT-4o, May 13th (temp=0)	100	100	100	100	38	87.6%
Claude 3.5 Sonnet	100	100	100	89	33	84.4%
Claude Sonnet 4	100	100	100	64	57	84.3%
Cohere Command R+ (Aug. 2024)	100	100	100	100	18	83.6%
Z.AI GLM 4.7 Flash	100	97	92	75	46	82.1%
Claude Opus 4	100	100	100	97	13	82.0%
Llama 3.1 8B	100	100	100	89	13	80.3%
Gemini 2.5 Flash Lite	100	100	92	57	50	79.8%
DeepSeek V3.1	100	100	100	72	26	79.7%
Llama 3.1 70B	100	100	100	89	0	77.8%
Ministral 3B	100	100	100	62	26	77.6%
GPT-4o, Aug. 6th (temp=1)	100	100	100	42	33	75.1%
Mistral Small 3.2 24B	100	100	100	67	0	73.3%
Ministral 3 8B	100	100	92	71	0	72.6%
Hermes 3 405B	100	100	100	62	0	72.4%
GPT-4o, May 13th (temp=1)	100	100	100	60	0	72.1%
Mistral NeMO	100	100	100	33	26	71.9%
DeepSeek V3 (2025-03-24)	100	100	100	40	0	68.0%
Claude 3.5 Haiku	100	100	100	33	0	66.7%
Rocinante 12B	100	100	46	46	40	66.5%
Claude 3.7 Sonnet	100	100	74	54	0	65.6%
Claude 3 Haiku	100	100	100	18	10	65.5%
Gemini 2.5 Flash	95	75	67	64	0	60.0%
Hermes 3 70B	100	100	97	0	0	59.5%
Minimax M2.5	93	67	63	62	8	58.5%
DeepSeek V3 (2024-12-26)	100	91	62	22	0	55.0%
Arcee AI: Trinity Mini	100	62	42	33	0	47.5%
GPT-4o Mini (temp=0)	100	80	18	18	10	45.1%
Claude Haiku 4.5	100	64	43	0	0	41.4%
Gemma 3 12B	100	43	21	21	0	37.2%
GPT-4.1 Mini	82	54	43	5	0	36.8%
Arcee AI: Trinity Large (Preview)	100	50	0	0	0	30.0%
GPT-4.1 Nano	100	42	0	0	0	28.4%

▼

Mystery: examining a crime scene

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
o4 Mini	100	100	100	100	100	100.0%
Z.AI GLM 5	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Claude Sonnet 4	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Z.AI GLM 4.7 Flash	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Grok 4 Fast	100	100	100	100	100	100.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
Claude 3.7 Sonnet	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
Z.AI GLM 4.5	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
DeepSeek V3.1	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
GPT-4o Mini (temp=1)	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Gemma 3 12B	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Llama 3.1 Nemotron 70B	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	95	98.9%
GPT-4.1 Mini	100	100	100	97	95	98.4%
Gemini 3 Pro (Preview)	100	100	100	100	87	97.4%
Gemini 2.5 Flash Lite	100	100	100	100	82	96.5%
Arcee AI: Trinity Mini	100	100	100	100	82	96.5%
Gemini 2.5 Flash	100	100	100	100	80	96.0%
Claude Opus 4.5	100	100	100	100	75	95.0%
DeepSeek-V2 Chat	100	100	100	97	75	94.5%
Gemma 3 4B	100	100	100	97	72	94.0%
Minimax M2.5	100	100	100	100	67	93.3%
Gemini 2.5 Pro	100	100	100	100	62	92.4%
WizardLM 2 8x22b	100	100	100	87	71	91.6%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	57	91.4%
Mistral Large	100	100	100	100	52	90.4%
Z.AI GLM 4.6	100	100	100	86	62	89.6%
GPT-4o, May 13th (temp=1)	100	100	100	100	46	89.2%
Gemini 3.1 Pro (Preview)	100	100	100	79	67	89.1%
DeepSeek V3 (2025-03-24)	100	100	100	100	40	88.0%
Claude 3.5 Sonnet	100	100	100	100	33	86.7%
Claude 3 Haiku	100	100	100	100	33	86.7%
Claude Sonnet 4.5	100	100	100	100	21	84.3%
GPT-4.1 Nano	100	100	89	89	29	81.3%
ByteDance Seed 1.6	100	100	100	100	0	80.0%
Claude 3.5 Haiku	100	100	100	100	0	80.0%
Hermes 3 405B	100	100	100	100	0	80.0%
Rocinante 12B	100	100	100	100	0	80.0%
Ministral 3 8B	100	100	95	67	33	78.9%
GPT-5 Nano	100	100	75	65	52	78.3%
Gemma 3 27B	100	100	95	42	5	68.3%
Ministral 3B	100	100	82	18	0	60.1%
Claude Haiku 4.5	100	100	67	26	0	58.6%
Hermes 3 70B	100	100	0	0	0	40.0%

▼

Romance: separated couple reunites

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Z.AI GLM 4.7	100	100	100	100	100	100.0%
Gemini 3 Pro (Preview)	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%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Z.AI GLM 4.6	100	100	100	100	100	100.0%
GPT-4.1	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
DeepSeek V3 (2025-03-24)	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
DeepSeek V3.2	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Mistral Medium 3.1	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Mistral Large	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%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Mistral NeMO	100	100	100	100	100	100.0%
WizardLM 2 8x22b	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
MoonshotAI: Kimi K2.5	100	100	100	100	95	98.9%
Claude Sonnet 4.6	100	100	100	100	95	98.9%
Grok 4 Fast	100	100	100	100	92	98.4%
DeepSeek V3.1	100	100	100	100	92	98.4%
Mistral Large 2	100	100	100	100	92	98.4%
Claude 3.5 Sonnet	100	100	100	100	89	97.8%
Claude 3.7 Sonnet	100	100	100	100	89	97.8%
Writer: Palmyra X5	100	100	100	100	89	97.8%
Gemma 3 4B	100	100	100	100	89	97.8%
Ministral 8B	100	100	100	100	86	97.1%
Claude Opus 4	100	100	100	100	85	97.0%
GPT-4o Mini (temp=1)	100	100	100	97	85	96.2%
Claude Haiku 4.5	100	100	100	100	80	96.0%
o4 Mini High	100	100	100	100	75	95.0%
GPT-5 Nano	100	100	100	100	73	94.6%
DeepSeek-V2 Chat	100	100	100	100	70	94.0%
o4 Mini	100	100	100	100	54	90.7%
Z.AI GLM 4.7 Flash	100	100	100	81	71	90.3%
Ministral 3 3B	100	100	100	100	40	88.0%
Z.AI GLM 4.5	100	100	100	100	31	86.1%
Gemini 2.5 Flash	100	100	100	100	30	86.0%
GPT-5 Mini	100	100	90	70	68	85.7%
Cohere Command R+ (Aug. 2024)	100	100	100	75	52	85.4%
Gemma 3 27B	100	100	100	87	18	81.0%
DeepSeek V3 (2024-12-26)	100	100	100	100	0	80.0%
Llama 3.1 Nemotron 70B	100	100	100	100	0	80.0%
Mistral Small 3.2 24B	100	100	100	100	0	80.0%
Minimax M2.5	100	100	100	57	39	79.2%
Z.AI GLM 5	100	100	100	57	36	78.6%
GPT-4o, Aug. 6th (temp=1)	100	100	100	46	40	77.2%
GPT-4.1 Mini	100	100	100	52	29	76.1%
Claude Opus 4.5	100	100	100	38	25	72.6%
Hermes 3 405B	100	100	100	52	0	70.4%
Gemini 2.5 Flash Lite	100	100	75	57	10	68.3%
Claude Sonnet 4	100	100	100	18	0	63.6%
Rocinante 12B	100	100	100	18	0	63.6%
Arcee AI: Trinity Large (Preview)	100	100	100	13	0	62.5%
GPT-4o, May 13th (temp=1)	100	100	95	0	0	58.9%
Gemma 3 12B	100	100	43	33	0	55.3%
GPT-4.1 Nano	100	79	71	0	0	50.0%
Claude 3 Haiku	100	100	33	14	0	49.5%
Arcee AI: Trinity Mini	100	89	40	0	0	45.8%
Hermes 3 70B	100	57	46	0	0	40.7%
Llama 3.1 8B	100	89	0	0	0	37.8%
Llama 3.1 70B	75	57	18	18	0	33.7%

▼

Thriller: chase through city streets

Model	# 1	# 2	# 3	# 4	# 5	Avg ▼
Gemini 3.1 Pro (Preview)	100	100	100	100	100	100.0%
Qwen 3.5 397B A17B	100	100	100	100	100	100.0%
GPT-5 Mini	100	100	100	100	100	100.0%
GPT-5.1	100	100	100	100	100	100.0%
o4 Mini High	100	100	100	100	100	100.0%
GPT-5	100	100	100	100	100	100.0%
Gemini 2.5 Pro	100	100	100	100	100	100.0%
GPT-5.2	100	100	100	100	100	100.0%
Claude Opus 4	100	100	100	100	100	100.0%
Grok 4	100	100	100	100	100	100.0%
Claude Sonnet 4.6	100	100	100	100	100	100.0%
Claude Sonnet 4.5	100	100	100	100	100	100.0%
Grok 4.1 Fast	100	100	100	100	100	100.0%
ByteDance Seed 1.6	100	100	100	100	100	100.0%
Gemini 3 Flash (Preview)	100	100	100	100	100	100.0%
Stealth: Aurora Alpha	100	100	100	100	100	100.0%
Qwen 3.5 Plus (2026-02-15)	100	100	100	100	100	100.0%
Claude 3.5 Sonnet	100	100	100	100	100	100.0%
Mistral Large 3	100	100	100	100	100	100.0%
GPT-4o, May 13th (temp=0)	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=1)	100	100	100	100	100	100.0%
GPT-4o, Aug. 6th (temp=0)	100	100	100	100	100	100.0%
Claude 3.5 Haiku	100	100	100	100	100	100.0%
ByteDance Seed 1.6 Flash	100	100	100	100	100	100.0%
Writer: Palmyra X5	100	100	100	100	100	100.0%
Mistral Large 2	100	100	100	100	100	100.0%
Gemini 2.5 Flash	100	100	100	100	100	100.0%
GPT-4o Mini (temp=0)	100	100	100	100	100	100.0%
Llama 3.1 70B	100	100	100	100	100	100.0%
Gemini 2.5 Flash Lite	100	100	100	100	100	100.0%
Mistral Large	100	100	100	100	100	100.0%
Ministral 3 14B	100	100	100	100	100	100.0%
Qwen 2.5 72B	100	100	100	100	100	100.0%
Mistral Small 3.2 24B	100	100	100	100	100	100.0%
Arcee AI: Trinity Large (Preview)	100	100	100	100	100	100.0%
Claude 3 Haiku	100	100	100	100	100	100.0%
Ministral 3 8B	100	100	100	100	100	100.0%
Arcee AI: Trinity Mini	100	100	100	100	100	100.0%
Ministral 3 3B	100	100	100	100	100	100.0%
Llama 3.1 8B	100	100	100	100	100	100.0%
Cohere Command R+ (Aug. 2024)	100	100	100	100	100	100.0%
Gemma 3 4B	100	100	100	100	100	100.0%
Ministral 8B	100	100	100	100	100	100.0%
Ministral 3B	100	100	100	100	100	100.0%
Rocinante 12B	100	100	100	100	100	100.0%
Claude Opus 4.6	100	100	100	100	95	98.9%
DeepSeek V3 (2024-12-26)	100	100	100	100	95	98.9%
Z.AI GLM 4.5	100	100	100	100	92	98.4%
o4 Mini	100	100	100	100	82	96.5%
Minimax M2.5	100	100	100	100	79	95.8%
GPT-4.1	100	100	100	100	75	95.0%
Claude Haiku 4.5	100	100	100	100	75	95.0%
Mistral NeMO	100	100	100	100	75	95.0%
Gemini 3 Pro (Preview)	100	100	100	100	67	93.3%
WizardLM 2 8x22b	100	100	100	89	70	91.7%
DeepSeek-V2 Chat	100	100	100	100	46	89.2%
Grok 4 Fast	100	100	100	100	33	86.7%
MoonshotAI: Kimi K2.5	100	100	100	100	33	86.7%
GPT-4o, May 13th (temp=1)	100	100	100	100	18	83.6%
Gemma 3 27B	100	100	100	95	10	80.9%
Z.AI GLM 4.7	100	100	100	57	46	80.7%
Llama 3.1 Nemotron 70B	100	100	89	57	57	80.6%
Claude Opus 4.5	100	100	100	100	0	80.0%
Claude Sonnet 4	100	100	100	100	0	80.0%
Z.AI GLM 4.6	100	100	100	100	0	80.0%
Z.AI GLM 4.7 Flash	100	100	100	100	0	80.0%
Claude 3.7 Sonnet	100	100	100	100	0	80.0%
DeepSeek V3.2	100	100	100	100	0	80.0%
Mistral Medium 3.1	100	100	100	100	0	80.0%
DeepSeek V3.1	100	100	100	100	0	80.0%
Gemma 3 12B	100	100	100	100	0	80.0%
Mistral Small Creative	100	100	100	100	0	80.0%
Z.AI GLM 5	100	100	100	89	0	77.8%
GPT-4.1 Mini	100	100	100	67	0	73.3%
GPT-4o Mini (temp=1)	100	100	100	67	0	73.3%
GPT-5 Nano	100	100	79	50	33	72.4%
Hermes 3 70B	100	100	100	38	18	71.2%
DeepSeek V3 (2025-03-24)	100	100	100	33	0	66.7%
GPT-4.1 Nano	100	100	75	57	0	66.4%
Hermes 3 405B	100	100	100	0	0	60.0%

Adverbs in dialogue tags

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

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

Novelcrafter Default Prompt

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