Skip to content

Benchmark

IdrockBench Uzbek translation uz-en/uz-ru (chrF++, task v2.0)

Multilingual benchmarks compare model performance across languages. Higher scores are better.

Models measured
12
Scores recorded
12
Sources
1
IQ weight
1.58

Score distribution

42.7
Mean
47.7
Median
12
Std dev
26.8
10th percentile
53.3
90th percentile
14.9
Min
14.954.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.58 for chat models, measured over 12 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 4 31B ITQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs54.6
2Gemma 4 26BQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs53.3
3Qwen3.6 27BQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs52
4Qwen3.5 27BQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs51.6
5Qwen 3.8 27BQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs51
6Qwen3.5 9BQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs47.7
7gemma 4 E4BQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs47.4
8Qwen3.5 4BQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs41.7
9Gemma 4 E2B ITQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs40.9
10Qwen3.5 2BQ8_0, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs30.1
11Nemotron 3.5 Lightning 30bQ4_K_M, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs26.8
12Qwen3.5 0.8BQ8_0, reasoning: off, organiser-run· IdrockBench Uzbek leaderboard data docs14.9

Data credits: IdrockBench, IDROCK AI Excellence Center, New Uzbekistan University. Public numeric results; no results licence stated (code MIT and dataset licences not borrowed).