Loading...
Loading...
NYT Connections
Twitter45
Total Models
32.68
Average Score
2.20 - 92.40
Score Range
1
Max Score Achievable
Top 10 Model Performance
Top 10 of 45
Models Using This Benchmark(45)
OpenAI(14 models)
o3 Pro
openai
87.30%
o1 pro
openai
82.50%
o3
openai
79.50%
o4 Mini
openai
74.70%
o1
openai
70.80%
o3-mini
openai
61.40%
GPT-4.5
openai
34.20%
o1 mini
openai
26.90%
GPT-4.1
openai
23.60%
GPT-4o
openai
18.70%
GPT-4o
openai
17.80%
GPT-4.1 Mini
openai
15.10%
GPT-4o-mini
openai
9.90%
GPT-4.1 Nano
openai
8.60%
Anthropic(7 models)
Claude Opus 4
anthropic
52.70%
Claude Sonnet 4
anthropic
41.40%
Claude 3.7 Sonnet
anthropic
33.60%
Claude 3 Opus
anthropic
19.20%
Claude 3.5 Sonnet
anthropic
17.70%
Claude 3.5 Haiku
anthropic
10.30%
Claude 3 Haiku
anthropic
2.20%
Google(5 models)
Gemini 2.5 Pro Preview
google
58.70%
54.10%
Gemini 2.5 Pro Preview
google
42.50%
Gemma 2 27B
google
12.20%
Gemma 3 27B
google
11.80%
DeepSeek(4 models)
Meta(4 models)
xAI(4 models)
Qwen(3 models)
Amazon(1 model)
Nova Pro 1.0
amazon
10.10%
Cohere(1 model)
Command A
cohere
13.60%
Microsoft(1 model)
Phi 4
microsoft
10.20%
Mistral(1 model)
Mistral Large 2
mistral
12.60%