Skip to content

Benchmark

ArmBench-LLM Armenian finer (ner_accuracy, 0-shot)

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
0.08

Score distribution

0.596
Mean
0.718
Median
0.237
Std dev
0.277
10th percentile
0.793
90th percentile
0
Min
00.806
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.08 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 docs0.806
2GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.806
3Claude Sonnet 4· ArmBench-LLM Armenian data docs0.799
4Claude Sonnet 3.7· ArmBench-LLM Armenian data docs0.793
5Mistral Large 3· ArmBench-LLM Armenian data docs0.792
6Gemini 3 Flash· ArmBench-LLM Armenian data docs0.777
7Gemini 2.5 Flash· ArmBench-LLM Armenian data docs0.769
8Qwen3.5 27B· ArmBench-LLM Armenian data docs0.769
9Ministral 14B· ArmBench-LLM Armenian data docs0.768
10Grok 4 Fast· ArmBench-LLM Armenian data docs0.761
11Grok 4.20· ArmBench-LLM Armenian data docs0.738
12Grok 4.1 Fast· ArmBench-LLM Armenian data docs0.736
13Gemma 4 31B IT· ArmBench-LLM Armenian data docs0.734
14GPT-5.4 mini· ArmBench-LLM Armenian data docs0.718
15Gemma 3 27B IT· ArmBench-LLM Armenian data docs0.703
16Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs0.702
17Gemma 3 12B IT· ArmBench-LLM Armenian data docs0.7
18GPT 5.4 Nano· ArmBench-LLM Armenian data docs0.675
19Gemma 3 4B IT· ArmBench-LLM Armenian data docs0.588
20Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs0.486
21Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs0.45
22Qwen3 4B Base· ArmBench-LLM Armenian data docs0.361
23GLM 4.7 Flash· ArmBench-LLM Armenian data docs0.349
24Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs0.317
25Gemini 2.5 Pro· ArmBench-LLM Armenian data docs0.277
26GLM 5· ArmBench-LLM Armenian data docs0.269
27Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs0.034
28Qwen3.5 9B· ArmBench-LLM Armenian data docs0

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