Skip to content

Benchmark

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

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
1.18

Score distribution

13.9
Mean
16
Median
8.5
Std dev
1.5
10th percentile
20.2
90th percentile
0
Min
031.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.18 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 docs31.1
2Gemma 4 31B IT· ArmBench-LLM Armenian data docs28.4
3Claude Sonnet 4· ArmBench-LLM Armenian data docs27.8
4GPT-5.4 mini· ArmBench-LLM Armenian data docs20.2
5Gemini 2.5 Flash· ArmBench-LLM Armenian data docs19.9
6GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs19.8
7GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs19.8
8Grok 4.1 Fast· ArmBench-LLM Armenian data docs19.4
9Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs18.9
10Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs18.6
11Grok 4 Fast· ArmBench-LLM Armenian data docs18.2
12Claude Sonnet 3.7· ArmBench-LLM Armenian data docs18.1
13Gemini 2.5 Pro· ArmBench-LLM Armenian data docs16.7
14GPT 5.4 Nano· ArmBench-LLM Armenian data docs16
15GLM 5· ArmBench-LLM Armenian data docs14.1
16GLM 4.7 Flash· ArmBench-LLM Armenian data docs13.8
17Gemma 3 27B IT· ArmBench-LLM Armenian data docs12.1
18Mistral Large 3· ArmBench-LLM Armenian data docs11.7
19Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs11.1
20Grok 4.20· ArmBench-LLM Armenian data docs9.3
21Gemma 3 4B IT· ArmBench-LLM Armenian data docs8.2
22Ministral 14B· ArmBench-LLM Armenian data docs8.2
23Gemma 3 12B IT· ArmBench-LLM Armenian data docs3.8
24Qwen3.5 27B· ArmBench-LLM Armenian data docs2.2
25Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs1.5
26Qwen3.5 9B· ArmBench-LLM Armenian data docs0.327
27Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs0.245
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.