Skip to content

Benchmark

ArmBench-LLM Armenian space_fix (space_accuracy, 0-shot)

Higher scores are better.

Models measured
27
Scores recorded
28
Sources
1
IQ weight
0.33

Score distribution

0.738
Mean
0.889
Median
0.297
Std dev
0.228
10th percentile
0.974
90th percentile
0
Min
00.984
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.33 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
1Claude Sonnet 3.7· ArmBench-LLM Armenian data docs0.984
2Gemini 3 Flash· ArmBench-LLM Armenian data docs0.977
3GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.974
4GPT-5.2 Prosetting not named by the source· ArmBench-LLM Armenian data docs0.974
5Gemini 2.5 Flash· ArmBench-LLM Armenian data docs0.966
6Gemma 4 31B IT· ArmBench-LLM Armenian data docs0.962
7GPT-5.4 mini· ArmBench-LLM Armenian data docs0.955
8Claude Sonnet 4· ArmBench-LLM Armenian data docs0.953
9Grok 4 Fast· ArmBench-LLM Armenian data docs0.94
10Qwen3.5 27B· ArmBench-LLM Armenian data docs0.934
11Grok 4.1 Fast· ArmBench-LLM Armenian data docs0.899
12Gemma 3 27B IT· ArmBench-LLM Armenian data docs0.895
13Mistral Large 3· ArmBench-LLM Armenian data docs0.893
14GPT 5.4 Nano· ArmBench-LLM Armenian data docs0.889
15Gemma 3 12B IT· ArmBench-LLM Armenian data docs0.848
16Mistral Small 3.2 24B Instruct· ArmBench-LLM Armenian data docs0.837
17Qwen3 4B Instruct 2507· ArmBench-LLM Armenian data docs0.806
18Ministral 14B· ArmBench-LLM Armenian data docs0.799
19Qwen3.5 35B A3B· ArmBench-LLM Armenian data docs0.786
20Gemini 3 Pro Preview· ArmBench-LLM Armenian data docs0.781
21Gemma 3 4B IT· ArmBench-LLM Armenian data docs0.76
22GLM 5· ArmBench-LLM Armenian data docs0.488
23GLM 4.7 Flash· ArmBench-LLM Armenian data docs0.469
24Gemini 2.5 Pro· ArmBench-LLM Armenian data docs0.459
25Gemini 3.1 Pro Preview· ArmBench-LLM Armenian data docs0.228
26Qwen3 4B Base· ArmBench-LLM Armenian data docs0.199
27Grok 4.20· ArmBench-LLM Armenian data docs0.00978
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.