Skip to content

Benchmark

Dubesor STEM

Scores are percentages.

Models measured
225
Scores recorded
268
Sources
1
IQ weight
1.64

Score distribution

51.4%
Mean
55.8%
Median
23.1%
Std dev
15.8%
10th percentile
79.9%
90th percentile
0.5%
Min
0.5%93.8%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.64 for chat models, measured over 225 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
201Gemini 2.0 Flash Lite· Dubesor LLM Benchmark34.9%
202GLM 4 32B· Dubesor LLM Benchmark34.9%
203Mistral Large2402· Dubesor LLM Benchmark34.9%
204Gemma 3 12B IT· Dubesor LLM Benchmark34.7%
205Llama 3 70B Instruct· Dubesor LLM Benchmark33.6%
206Mistral Medium· Dubesor LLM Benchmark33.1%
207Llama 3.2 90B Vision Instruct· Dubesor LLM Benchmark32.6%
208gemini-1.5-flash-002· Dubesor LLM Benchmark32.5%
209Qwen2.5 32B Instruct· Dubesor LLM Benchmark31.5%
210Hunyuan A13B Instructthinking· Dubesor LLM Benchmark31.1%
211jamba-1.6-large· Dubesor LLM Benchmark30.5%
212ERNIE 4.5 21B A3B· Dubesor LLM Benchmark30.2%
213Llama 3.1 70B Instruct· Dubesor LLM Benchmark30.1%
214Gemini 1.5 Flash· Dubesor LLM Benchmark28.8%
215GPT-4.1 nano· Dubesor LLM Benchmark28.8%
216Claude 2.1· Dubesor LLM Benchmark28.6%
217Inflection 3 Productivity· Dubesor LLM Benchmark27.8%
218Qwen3 Coder 30B A3B Instruct· Dubesor LLM Benchmark27.6%
219Jamba 1.5 Large· Dubesor LLM Benchmark27.5%
220GLM 4· Dubesor LLM Benchmark27.2%
221Nemotron 3 Nano 30B A3B· Dubesor LLM Benchmark27.1%
222Ministral 3 14B Instruct· Dubesor LLM Benchmark26.7%
223Olmo 3 7B Instruct· Dubesor LLM Benchmark26.3%
224Hunyuan A13B Instruct· Dubesor LLM Benchmark24.9%
225Gemma 2 27B IT· Dubesor LLM Benchmark24.8%
226Inflection 3 Pi· Dubesor LLM Benchmark24.8%
227GPT-3.5 Turbo· Dubesor LLM Benchmark24%
228Gemini 1.5 Flash-8B· Dubesor LLM Benchmark22.8%
229Gemma 4 E2B-itthinking· Dubesor LLM Benchmark22.3%
230Qwen3 4B· Dubesor LLM Benchmark22.2%
231Gemma 3n E4B IT· Dubesor LLM Benchmark21.6%
232Ministral 3 8b Instruct· Dubesor LLM Benchmark21.3%
233Qwen 2.5 14B Instruct· Dubesor LLM Benchmark20.6%
234Command R 08 2024· Dubesor LLM Benchmark20.3%
235Qwen3.5 4B· Dubesor LLM Benchmark19.1%
236Qwen 2.5 Coder 32B Instruct· Dubesor LLM Benchmark18.3%
237Phi-3 Medium 128K Instruct· Dubesor LLM Benchmark17.6%
238Gemini Pro· Dubesor LLM Benchmark17.1%
239Qwen3 1.7Bthinking· Dubesor LLM Benchmark17%
240Llama 2 70B Chat· Dubesor LLM Benchmark16%
241Qwen2.5 Coder 7B Instruct· Dubesor LLM Benchmark15.8%
242Qwen2.5 7B Instruct· Dubesor LLM Benchmark15.5%
243Mixtral 8x7B Instruct v0.1· Dubesor LLM Benchmark14.2%
244Olmo 2 32B Instruct· Dubesor LLM Benchmark13.2%
245Phi-3 Mini 4K Instruct· Dubesor LLM Benchmark13.1%
246Reka Core· Dubesor LLM Benchmark13%
247Command R7b 12 2024· Dubesor LLM Benchmark12.8%
248LFM-7b· Dubesor LLM Benchmark11.8%
249Command R+· Dubesor LLM Benchmark11.5%
250Mistral Nemo 12B Instruct· Dubesor LLM Benchmark11.5%

Showing 201–250 of 268 results.