Skip to content

Benchmark

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

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
0.00

Score distribution

17.5
Mean
18
Median
9.5
Std dev
4.4
10th percentile
29.3
90th percentile
0.000901
Min
0.00090134.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.00 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
1Gemma 3 12B IT· ArmBench-LLM Armenian data docs34.5
2Gemma 3 4B IT· ArmBench-LLM Armenian data docs30.4
3Grok 4.1 Fast· ArmBench-LLM Armenian data docs29.9
4Qwen3.5 27B· ArmBench-LLM Armenian data docs29.3
5Gemini 2.5 Flash· ArmBench-LLM Armenian data docs28
6Gemma 3 27B IT· ArmBench-LLM Armenian data docs26.9
7Gemma 4 31B IT· ArmBench-LLM Armenian data docs26.3
8GLM 5· ArmBench-LLM Armenian data docs25.2
9GLM 4.7 Flash· ArmBench-LLM Armenian data docs25.2
10Qwen3.5 9B· ArmBench-LLM Armenian data docs23.9
11Grok 4 Fast· ArmBench-LLM Armenian data docs23.5
12Claude Sonnet 3.7· ArmBench-LLM Armenian data docs18.4
13Gemini 2.5 Pro· ArmBench-LLM Armenian data docs18.2
14Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs18
15Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs17.9
16GPT-5.4 mini· ArmBench-LLM Armenian data docs16.5
17Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs15.4
18GPT 5.4 Nano· ArmBench-LLM Armenian data docs15.2
19Claude Sonnet 4· ArmBench-LLM Armenian data docs11.8
20Gemini 3 Flash· ArmBench-LLM Armenian data docs11.3
21GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs10.6
22GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs10.6
23Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs7.1
24Mistral Large 3· ArmBench-LLM Armenian data docs6.8
25Qwen3 4B Base· ArmBench-LLM Armenian data docs4.4
26Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs3.4
27Ministral 14B· ArmBench-LLM Armenian data docs2.3
28Grok 4.20· ArmBench-LLM Armenian data docs0.000901

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