Skip to content

Benchmark

Dubesor Reason

Scores are percentages.

Models measured
218
Scores recorded
261
Sources
1
IQ weight
1.50

Score distribution

44.7%
Mean
42.8%
Median
24.3%
Std dev
11.2%
10th percentile
76.9%
90th percentile
0%
Min
0%93.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.50 for chat models, measured over 218 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101gemini-2.5-flash-preview-04-17· Dubesor LLM Benchmark53%
102DeepSeek V3.1· Dubesor LLM Benchmark52.9%
103Llama 3.3 Nemotron Super 49B v1.5thinking· Dubesor LLM Benchmark52.8%
104Llama 3.3 70B Instruct· Dubesor LLM Benchmark52.5%
105Llama 3.3 Nemotron Super 49B v1· Dubesor LLM Benchmark52.4%
106Qwen3.5 35B A3B· Dubesor LLM Benchmark51.7%
107Llama 4 Maverick· Dubesor LLM Benchmark51.5%
108Phi 4 Reasoning Plus· Dubesor LLM Benchmark51.3%
109r1-1776· Dubesor LLM Benchmark51.3%
110Claude Haiku 4.5· Dubesor LLM Benchmark50%
111Qwen3 Coder Next· Dubesor LLM Benchmark49.6%
112Llama 3.1 Nemotron Ultra 253B v1thinking· Dubesor LLM Benchmark49.4%
113Magistral Medium2506· Dubesor LLM Benchmark49.2%
114Gemini 2.5 Flash Preview 05-20 (no think)· Dubesor LLM Benchmark49.1%
115Mistral Medium 3.1· Dubesor LLM Benchmark49%
116Llama 3.3 Nemotron Super 49B v1thinking· Dubesor LLM Benchmark48.9%
117Qwen3 VL 32B Instruct· Dubesor LLM Benchmark48.3%
118LongCat Flash Chat· Dubesor LLM Benchmark47.9%
119qwen3-vl-30b· Dubesor LLM Benchmark47.7%
120Gemini 2.5 Flash· Dubesor LLM Benchmark47.3%
121Grok 4.20non-reasoning· Dubesor LLM Benchmark46.4%
122Mistral Large 3· Dubesor LLM Benchmark46.3%
123Qwen3 32B· Dubesor LLM Benchmark46.3%
124GPT-4.1 mini· Dubesor LLM Benchmark44.2%
125Llama 3.3 Nemotron Super 49B v1.5· Dubesor LLM Benchmark44.2%
126MiniMax M2· Dubesor LLM Benchmark44.1%
127QwQ 32B· Dubesor LLM Benchmark43.8%
128Qwen2.5-Max· Dubesor LLM Benchmark43.4%
129Qwen3 32Bthinking· Dubesor LLM Benchmark43.4%
130Devstral 2 123b Instruct 2512· Dubesor LLM Benchmark43.1%
131Qwen2.5 72B Instruct· Dubesor LLM Benchmark42.8%
132Mistral Large2402· Dubesor LLM Benchmark42.2%
133Qwen3 14Bthinking· Dubesor LLM Benchmark42%
134Olmo 3 32B Think· Dubesor LLM Benchmark41.9%
135Qwen3 8Bthinking· Dubesor LLM Benchmark41.8%
136GLM 4.5 Airthinking· Dubesor LLM Benchmark41.7%
137Ling 1t· Dubesor LLM Benchmark41.7%
138Llama 3.1 Nemotron 70B Instruct· Dubesor LLM Benchmark41.6%
139Mistral Medium 3· Dubesor LLM Benchmark41.5%
140gemini-1.5-flash-002· Dubesor LLM Benchmark41.2%
141gemini-1.5-pro-002· Dubesor LLM Benchmark40.9%
142Gemini 2.0 Flash· Dubesor LLM Benchmark40.8%
143GLM 4.6vthinking· Dubesor LLM Benchmark40.6%
144MiniMax M1· Dubesor LLM Benchmark40.6%
145Mistral Large 2· Dubesor LLM Benchmark40.5%
146DeepSeek V3· Dubesor LLM Benchmark40.3%
147yi-large· Dubesor LLM Benchmark40.2%
148Llama 3.2 90B Vision Instruct· Dubesor LLM Benchmark39.9%
149Gemini 1.5 Pro· Dubesor LLM Benchmark39.7%
150Nemotron-4 340B Instruct· Dubesor LLM Benchmark39.7%

Showing 101–150 of 261 results.