Skip to content

Benchmark

ArmBench-LLM Armenian arak (bleu, 0-shot)

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
0.16

Score distribution

1.1
Mean
1.1
Median
0.744
Std dev
0.033
10th percentile
2.1
90th percentile
0.0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000427
Min
0.00000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000004272.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.16 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 Pro Preview· ArmBench-LLM Armenian data docs2.6
2Mistral Large 3· ArmBench-LLM Armenian data docs2.2
3Grok 4 Fast· ArmBench-LLM Armenian data docs2.1
4Grok 4.1 Fast· ArmBench-LLM Armenian data docs2.1
5Gemini 3 Flash· ArmBench-LLM Armenian data docs2.1
6Ministral 14B· ArmBench-LLM Armenian data docs1.8
7GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs1.8
8GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs1.8
9Qwen3.5 27B· ArmBench-LLM Armenian data docs1.6
10Gemini 2.5 Flash· ArmBench-LLM Armenian data docs1.5
11Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs1.4
12Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs1.2
13Gemma 3 27B IT· ArmBench-LLM Armenian data docs1.2
14Gemma 3 4B IT· ArmBench-LLM Armenian data docs1.1
15GLM 5· ArmBench-LLM Armenian data docs1.1
16GPT 5.4 Nano· ArmBench-LLM Armenian data docs1
17Gemma 4 31B IT· ArmBench-LLM Armenian data docs1
18Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs0.917
19Gemma 3 12B IT· ArmBench-LLM Armenian data docs0.875
20GLM 4.7 Flash· ArmBench-LLM Armenian data docs0.815
21GPT-5.4 mini· ArmBench-LLM Armenian data docs0.764
22Qwen3 4B Base· ArmBench-LLM Armenian data docs0.327
23Claude Sonnet 3.7· ArmBench-LLM Armenian data docs0.325
24Claude Sonnet 4· ArmBench-LLM Armenian data docs0.217
25Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs0.0332
26Gemini 2.5 Pro· ArmBench-LLM Armenian data docs0.00059
27Qwen3.5 9B· ArmBench-LLM Armenian data docs0.00000000000000000671
28Grok 4.20· ArmBench-LLM Armenian data docs0.0000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000427

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