Skip to content

Benchmark

Dubesor Total

Scores are percentages.

Models measured
227
Scores recorded
270
Sources
1
IQ weight
1.57

Score distribution

48.3%
Mean
50.7%
Median
20.5%
Std dev
18.3%
10th percentile
73.4%
90th percentile
0.2%
Min
0.2%86.1%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.57 for chat models, measured over 227 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101Claude Haiku 4.5· Dubesor LLM Benchmark56.6%
102GPT-5.1 Codex Mini· Dubesor LLM Benchmark56.5%
103Llama 3.1 Nemotron 70B Instruct· Dubesor LLM Benchmark56.4%
104LongCat Flash Chat· Dubesor LLM Benchmark56.3%
105Nemotron 3 Super 120B A12Bthinking· Dubesor LLM Benchmark56.1%
106Nemotron 3 Super 120B A12B· Dubesor LLM Benchmark56%
107Llama 4 Maverick· Dubesor LLM Benchmark55.9%
108Magistral Medium2506· Dubesor LLM Benchmark55.9%
109Qwen3.5 35B A3Bthinking· Dubesor LLM Benchmark55.6%
110Mistral Large 2· Dubesor LLM Benchmark55.4%
111Gemini 2.5 Flash Preview 05-20 (no think)· Dubesor LLM Benchmark54.9%
112Llama 3.3 Nemotron Super 49B v1· Dubesor LLM Benchmark54.8%
113gemini-1.5-pro-002· Dubesor LLM Benchmark54.7%
114Qwen3.5 27B· Dubesor LLM Benchmark54.6%
115Devstral 2 123b Instruct 2512· Dubesor LLM Benchmark54.5%
116Gemini 2.5 Flash· Dubesor LLM Benchmark54.2%
117Mistral Medium 3.1· Dubesor LLM Benchmark53.5%
118Qwen3 30B A3B Instruct· Dubesor LLM Benchmark53.3%
119Phi 4 Reasoning Plus· Dubesor LLM Benchmark53.2%
120Qwen3 32Bthinking· Dubesor LLM Benchmark53%
121Mistral Medium 3· Dubesor LLM Benchmark52.9%
122Gemini 2.0 Flash· Dubesor LLM Benchmark52.8%
123GPT-4.1 mini· Dubesor LLM Benchmark52.8%
124Qwen3 30B A3B· Dubesor LLM Benchmark52.6%
125GLM 4.6vthinking· Dubesor LLM Benchmark52.5%
126Grok Code Fast 1· Dubesor LLM Benchmark52.4%
127MiniMax M1· Dubesor LLM Benchmark52.3%
128Llama 3.3 Nemotron Super 49B v1.5· Dubesor LLM Benchmark52.2%
129GLM 4.5 Airthinking· Dubesor LLM Benchmark51.9%
130Qwen2.5-Max· Dubesor LLM Benchmark51.6%
131MiniMax M2· Dubesor LLM Benchmark51.4%
132Qwen3 Coder Plus· Dubesor LLM Benchmark51.3%
133Ling 1t· Dubesor LLM Benchmark50.9%
134Llama 3.1 Nemotron Ultra 253B v1· Dubesor LLM Benchmark50.9%
135GPT-5.4 mini· Dubesor LLM Benchmark50.7%
136Llama 3.1 Nemotron Ultra 253B v1thinking· Dubesor LLM Benchmark50.7%
137GPT-4o mini· Dubesor LLM Benchmark50.3%
138DeepSeek V2.5· Dubesor LLM Benchmark50%
139Qwen3 Coder Next· Dubesor LLM Benchmark49.7%
140Qwen3.5 35B A3B· Dubesor LLM Benchmark49.4%
141MiMo V2 Flash· Dubesor LLM Benchmark49.3%
142DeepSeek V2 Chat· Dubesor LLM Benchmark49.2%
143Llama 3.2 90B Vision Instruct· Dubesor LLM Benchmark49.2%
144Qwen3.5 9Bthinking· Dubesor LLM Benchmark48.9%
145Qwen3 30B A3Bthinking· Dubesor LLM Benchmark48.8%
146Qwen3 8Bthinking· Dubesor LLM Benchmark48.6%
147Qwen3 14Bthinking· Dubesor LLM Benchmark48.3%
148GPT OSS 120B· Dubesor LLM Benchmark48.1%
149Qwen2.5-Plus· Dubesor LLM Benchmark48.1%
150Gemini 2.5 Flash-Lite· Dubesor LLM Benchmark47.5%

Showing 101–150 of 270 results.