Skip to content

Benchmark

ArmBench-LLM Armenian exam_literature (armenian_exam_score, 0-shot)

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
1.10

Score distribution

7.2
Mean
7.5
Median
2.7
Std dev
4.5
10th percentile
10
90th percentile
0.75
Min
0.7512.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.10 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 docs12.5
2Gemini 2.5 Pro· ArmBench-LLM Armenian data docs11.8
3Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs11
4Claude Sonnet 3.7· ArmBench-LLM Armenian data docs10
5GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs10
6GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs10
7Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs9
8Grok 4 Fast· ArmBench-LLM Armenian data docs9
9Gemini 2.5 Flash· ArmBench-LLM Armenian data docs8.3
10Qwen3.5 27B· ArmBench-LLM Armenian data docs8.3
11Claude Sonnet 4· ArmBench-LLM Armenian data docs8
12Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs8
13Gemma 3 4B IT· ArmBench-LLM Armenian data docs7.5
14Grok 4.1 Fast· ArmBench-LLM Armenian data docs7.5
15Mistral Large 3· ArmBench-LLM Armenian data docs7.5
16Gemma 4 31B IT· ArmBench-LLM Armenian data docs7.3
17Gemma 3 12B IT· ArmBench-LLM Armenian data docs6.5
18GPT 5.4 Nano· ArmBench-LLM Armenian data docs6.3
19Grok 4.20· ArmBench-LLM Armenian data docs6.3
20Gemma 3 27B IT· ArmBench-LLM Armenian data docs5.8
21GLM 5· ArmBench-LLM Armenian data docs5.5
22GPT-5.4 mini· ArmBench-LLM Armenian data docs5
23Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs5
24Qwen3 4B Base· ArmBench-LLM Armenian data docs4.8
25Ministral 14B· ArmBench-LLM Armenian data docs4.5
26GLM 4.7 Flash· ArmBench-LLM Armenian data docs3.8
27Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs2.8
28Qwen3.5 9B· ArmBench-LLM Armenian data docs0.75

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