Skip to content

Benchmark

Dubesor Utility

Scores are percentages.

Models measured
224
Scores recorded
266
Sources
1
IQ weight
1.16

Score distribution

53.7%
Mean
55%
Median
17.9%
Std dev
28.7%
10th percentile
76.8%
90th percentile
4%
Min
4%87.4%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.16 for chat models, measured over 224 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Mistral Nemo 12B Instruct· Dubesor LLM Benchmark41.8%
202Qwen3 32Bthinking· Dubesor LLM Benchmark41.8%
203QwQ 32B· Dubesor LLM Benchmark41.8%
204GPT-5.4 mini· Dubesor LLM Benchmark40.2%
205Qwen3.5 9B· Dubesor LLM Benchmark40.1%
206Qwen3.5 27B· Dubesor LLM Benchmark39.4%
207Command R 08 2024· Dubesor LLM Benchmark39%
208Grok 2 mini· Dubesor LLM Benchmark38.4%
209Qwen3 4B Instruct 2507· Dubesor LLM Benchmark38.4%
210Hunyuan A13B Instructthinking· Dubesor LLM Benchmark38.1%
211GLM 4.7 Flashthinking· Dubesor LLM Benchmark37.8%
212GLM Z1 32B 0414· Dubesor LLM Benchmark37.6%
213Qwen3 4Bthinking· Dubesor LLM Benchmark37.6%
214ERNIE 4.5 300B A47B· Dubesor LLM Benchmark37.2%
215Gemma 4 E2B-itthinking· Dubesor LLM Benchmark37.2%
216Gemma 4 E4B-it· Dubesor LLM Benchmark37.2%
217Gemma 4 E4B-itthinking· Dubesor LLM Benchmark37.2%
218Ministral 3 14B Instruct· Dubesor LLM Benchmark37.2%
219Olmo 3 7B Think· Dubesor LLM Benchmark37.2%
220Qwen3 32B· Dubesor LLM Benchmark37%
221Nemotron-4 340B Instruct· Dubesor LLM Benchmark36.3%
222Qwen3.5 4Bthinking· Dubesor LLM Benchmark36.1%
223Gemma 3 4B IT· Dubesor LLM Benchmark35.8%
224Qwen3 Coder 30B A3B Instruct· Dubesor LLM Benchmark35.7%
225DeepSeek R1 0528 Qwen3 8B· Dubesor LLM Benchmark35.1%
226Qwen3 30B A3B Instruct· Dubesor LLM Benchmark34.4%
227GLM 4.7 Flash· Dubesor LLM Benchmark33.6%
228Gemini Pro· Dubesor LLM Benchmark33.4%
229LFM2-8B-A1B· Dubesor LLM Benchmark33.3%
230Qwen3 8B· Dubesor LLM Benchmark33.3%
231Qwen2 72B Instruct· Dubesor LLM Benchmark32.5%
232Gemma 4 E2B-it· Dubesor LLM Benchmark32.4%
233Qwen2.5 72B Instruct· Dubesor LLM Benchmark31.2%
234Qwen3 Coder Flash· Dubesor LLM Benchmark30.7%
235Grok 4 Fastreasoning· Dubesor LLM Benchmark30.3%
236Inflection 3 Pi· Dubesor LLM Benchmark29.4%
237ERNIE 4.5 21B A3B· Dubesor LLM Benchmark29%
238Qwen 2.5 14B Instruct· Dubesor LLM Benchmark28.7%
239Qwen 2.5 Coder 32B Instruct· Dubesor LLM Benchmark28.7%
240Magistral Small 2509· Dubesor LLM Benchmark28.6%
241MiniMax 01· Dubesor LLM Benchmark28.4%
242jamba-1.6-mini· Dubesor LLM Benchmark28.3%
243MythoMax L2 13b· Dubesor LLM Benchmark28.3%
244Qwen3.5 4B· Dubesor LLM Benchmark28.3%
245WizardLM 2 8x22B· Dubesor LLM Benchmark28%
246Ministral 3 8b Instruct· Dubesor LLM Benchmark27.5%
247GPT-5.2 Chat· Dubesor LLM Benchmark26%
248Command R7b 12 2024· Dubesor LLM Benchmark24.3%
249GPT-5 nano· Dubesor LLM Benchmark24%
250Qwen3 4B· Dubesor LLM Benchmark23.7%

Showing 201–250 of 266 results.