Skip to content

Benchmark

ArmBench-LLM Armenian mmlu_pro (armenian_mmlu_pro_score, 0-shot)

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
1.01

Score distribution

0.608
Mean
0.695
Median
0.238
Std dev
0.186
10th percentile
0.817
90th percentile
0
Min
00.868
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.01 for chat models, measured over 27 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Gemini 3 Flash· ArmBench-LLM Armenian data docs0.868
2GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.849
3GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.849
4Grok 4 Fast· ArmBench-LLM Armenian data docs0.817
5Grok 4.1 Fast· ArmBench-LLM Armenian data docs0.817
6Qwen3.5 27B· ArmBench-LLM Armenian data docs0.813
7Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs0.808
8Gemma 4 31B IT· ArmBench-LLM Armenian data docs0.795
9Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs0.751
10Claude Sonnet 4· ArmBench-LLM Armenian data docs0.748
11Claude Sonnet 3.7· ArmBench-LLM Armenian data docs0.746
12Mistral Large 3· ArmBench-LLM Armenian data docs0.741
13GPT-5.4 mini· ArmBench-LLM Armenian data docs0.705
14GLM 5· ArmBench-LLM Armenian data docs0.695
15Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs0.653
16Gemini 2.5 Flash· ArmBench-LLM Armenian data docs0.642
17Ministral 14B· ArmBench-LLM Armenian data docs0.626
18Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs0.606
19GLM 4.7 Flash· ArmBench-LLM Armenian data docs0.586
20Gemma 3 27B IT· ArmBench-LLM Armenian data docs0.584
21GPT 5.4 Nano· ArmBench-LLM Armenian data docs0.542
22Gemma 3 12B IT· ArmBench-LLM Armenian data docs0.506
23Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs0.496
24Gemma 3 4B IT· ArmBench-LLM Armenian data docs0.331
25Qwen3.5 9B· ArmBench-LLM Armenian data docs0.186
26Gemini 2.5 Pro· ArmBench-LLM Armenian data docs0.18
27Qwen3 4B Base· ArmBench-LLM Armenian data docs0.0931
28Grok 4.20· ArmBench-LLM Armenian data docs0

Data credits: ArmBench-LLM by Metric-AI. Public numeric primary task results; no separate results licence asserted.