Skip to content

Benchmark

MMLU

Scores are percentages. Higher is better.

Models and systems measured
138
Scores recorded
153
Sources
3
IQ weight
0.00

Score distribution

76.9%
Mean
80.1%
Median
12.2%
Std dev
60.1%
10th percentile
88.8%
90th percentile
36.6%
Min
36.6%92.5%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.00 for chat models, measured over 133 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#Model or systemScore
101Qwen2.5 7B Instruct· Epoch AI Benchmarking Hub data72.9%
102Gemma 2 9B IT· Epoch AI Benchmarking Hub data
Values from 2 sources
72.1%
103gemini-1.0-pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim71.8%
104Nova Microv1· Epoch AI Benchmarking Hub data70.8%
105Mixtral 8X7b V0.1· Epoch AI Benchmarking Hub data70.6%
106Qwen2 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim70.5%
107Text Davinci 002· Epoch AI Benchmarking Hub data70%
108GPT-3.5 Turbo· LLM Stats data
Values from 2 sources
69.8%
109Jamba 1.5 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.7%
110Llama 3.1 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim
Values from 2 sources
69.4%
111Pixtral 12B2409· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim69.2%
112Llama 3 8B Instruct· Epoch AI Benchmarking Hub data68.8%
113Mistral Large2402· Epoch AI Benchmarking Hub data68.8%
114Phi-3 Mini 4K Instruct· Epoch AI Benchmarking Hub data68.8%
115Mistral Small· Epoch AI Benchmarking Hub data68.7%
116Qwen1.5-14B· Epoch AI Benchmarking Hub data68.6%
117Mistral Nemo· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim68%
118Qwen2.5-Coder-7B· Epoch AI Benchmarking Hub data68%
119Qwen2.5 Coder 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.6%
120Phi 4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim67.3%
121gemma-7b· Epoch AI Benchmarking Hub data66.1%
122C4AI Command R 08 2024· Epoch AI Benchmarking Hub data65.2%
123Qwen-14B-Chat· Epoch AI Benchmarking Hub data65%
124Gemma 3n E4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim64.9%
125Starcoder2 15b· Epoch AI Benchmarking Hub data64.1%
126Yi-6B· Epoch AI Benchmarking Hub data64%
127Llama 3.2 3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim63.4%
128Qwen1.5-7B· Epoch AI Benchmarking Hub data62.6%
129Mistral 7B Instruct v0.2· Epoch AI Benchmarking Hub data62.5%
130Mistral 7B v0.1· Epoch AI Benchmarking Hub data62.5%
131deepseek-coder-v2-lite-base· Epoch AI Benchmarking Hub data60.5%
132Gemma 3n E2b It· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim60.1%
133Llama 2 70B Chat· Epoch AI Benchmarking Hub data59.9%
134Mistral (7B) Instruct v0.3· Epoch AI Benchmarking Hub data59.9%
135phi-2· Epoch AI Benchmarking Hub data58.4%
136Llama 3.2 11B Vision Instruct· Epoch AI Benchmarking Hub data56.5%
137Llama 2 13b· Epoch AI Benchmarking Hub data55.6%
138qwen2p5-coder-1p5b· Epoch AI Benchmarking Hub data53.6%
139Llama 2 13B Chat· Epoch AI Benchmarking Hub data50.9%
140ChatGLM2 6B· Epoch AI Benchmarking Hub data47.9%
141LLaMA 13B· Epoch AI Benchmarking Hub data47.7%
142Llama 2 7b· Epoch AI Benchmarking Hub data45.8%
143text-davinci-001· Epoch AI Benchmarking Hub data43.9%
144gemma-2b· Epoch AI Benchmarking Hub data42.3%
145qwen2p5-coder-0p5b· Epoch AI Benchmarking Hub data42%
146ERNIE 4.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim41.9%
147starcoder2-7b· Epoch AI Benchmarking Hub data38.8%
148starcoder2-3b· Epoch AI Benchmarking Hub data36.6%

Showing 101–148 of 148 results.

Not ranked

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

Model or systemScore
MiMo V2 Flash Baseofficial Base table · 5-shot· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.86.7%
Qwen3.5-397B-A17B-BaseQwen3.5 release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.88.6%
Qwen3.7-Plus-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.90.4%
Qwen3.8-27B-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.87.5%
Qwen3.8-Flash-Next-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.90.4%

Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.