Skip to content

Benchmark

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

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
1.26

Score distribution

0.646
Mean
0.66
Median
0.226
Std dev
0.38
10th percentile
0.92
90th percentile
0
Min
00.96
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.26 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 docs0.96
2GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.96
3Claude Sonnet 3.7· ArmBench-LLM Armenian data docs0.92
4Claude Sonnet 4· ArmBench-LLM Armenian data docs0.92
5Grok 4 Fast· ArmBench-LLM Armenian data docs0.92
6GPT-5.4 mini· ArmBench-LLM Armenian data docs0.9
7Gemini 2.5 Pro· ArmBench-LLM Armenian data docs0.82
8Qwen3.5 27B· ArmBench-LLM Armenian data docs0.8
9Grok 4.1 Fast· ArmBench-LLM Armenian data docs0.76
10Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs0.72
11Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs0.72
12GLM 4.7 Flash· ArmBench-LLM Armenian data docs0.7
13Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs0.68
14Gemini 3 Flash· ArmBench-LLM Armenian data docs0.66
15GLM 5· ArmBench-LLM Armenian data docs0.66
16Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs0.64
17Gemma 4 31B IT· ArmBench-LLM Armenian data docs0.62
18GPT 5.4 Nano· ArmBench-LLM Armenian data docs0.62
19Gemini 2.5 Flash· ArmBench-LLM Armenian data docs0.6
20Ministral 14B· ArmBench-LLM Armenian data docs0.58
21Mistral Large 3· ArmBench-LLM Armenian data docs0.54
22Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs0.5
23Qwen3.5 9B· ArmBench-LLM Armenian data docs0.5
24Gemma 3 27B IT· ArmBench-LLM Armenian data docs0.46
25Gemma 3 4B IT· ArmBench-LLM Armenian data docs0.38
26Gemma 3 12B IT· ArmBench-LLM Armenian data docs0.36
27Grok 4.20· ArmBench-LLM Armenian data docs0.2
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.