Skip to content

Benchmark

ArmBench-LLM Armenian hartak (exact_match_mcqa, 0-shot)

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
1.38

Score distribution

0.721
Mean
0.8
Median
0.226
Std dev
0.444
10th percentile
0.956
90th percentile
0
Min
00.978
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.38 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
1GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.978
2GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.978
3GPT-5.4 mini· ArmBench-LLM Armenian data docs0.956
4Grok 4 Fast· ArmBench-LLM Armenian data docs0.956
5Claude Sonnet 3.7· ArmBench-LLM Armenian data docs0.911
6GPT 5.4 Nano· ArmBench-LLM Armenian data docs0.911
7Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs0.889
8Gemma 4 31B IT· ArmBench-LLM Armenian data docs0.867
9Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs0.867
10GLM 4.7 Flash· ArmBench-LLM Armenian data docs0.844
11Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs0.844
12Gemini 2.5 Pro· ArmBench-LLM Armenian data docs0.822
13Qwen3.5 27B· ArmBench-LLM Armenian data docs0.822
14Gemini 3 Flash· ArmBench-LLM Armenian data docs0.8
15Grok 4.1 Fast· ArmBench-LLM Armenian data docs0.8
16Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs0.756
17GLM 5· ArmBench-LLM Armenian data docs0.711
18Mistral Large 3· ArmBench-LLM Armenian data docs0.711
19Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs0.689
20Grok 4.20· ArmBench-LLM Armenian data docs0.644
21Claude Sonnet 4· ArmBench-LLM Armenian data docs0.6
22Gemma 3 4B IT· ArmBench-LLM Armenian data docs0.6
23Qwen3.5 9B· ArmBench-LLM Armenian data docs0.6
24Gemma 3 12B IT· ArmBench-LLM Armenian data docs0.467
25Gemini 2.5 Flash· ArmBench-LLM Armenian data docs0.444
26Ministral 14B· ArmBench-LLM Armenian data docs0.4
27Gemma 3 27B IT· ArmBench-LLM Armenian data docs0.311
28Qwen3 4B Base· ArmBench-LLM Armenian data docs0

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