Skip to content

Benchmark

UzLiB correct word accuracy

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

Models measured
74
Scores recorded
74
Sources
1
IQ weight
1.32

Score distribution

0.572
Mean
0.566
Median
0.15
Std dev
0.38
10th percentile
0.794
90th percentile
0.229
Min
0.2290.845
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.32 for chat models, measured over 73 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 4 Carbontemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.845
2Gemini 4 Argontemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.841
3Gemini 4 Neontemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.838
4Gemini 3 Pro Previewtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.822
5Gemini 4 Hydrogentemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.814
6GPT-6 Lunatemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.81
7Gemini 3.7 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.806
8Gemini 3 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.794
9Claude Sonnet 5.5temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.786
10Muse Spark 1.3temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.784
11Gemini 3.1 Pro Previewtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.78
12Claude Opus 5.5temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.767
13Gemini 3.5 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.759
14DeepSeek V4.1 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.755
15GPT-5.6 Lunatemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.728
16DeepSeek V4 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.722
17Gemini 2.5 Protemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.68
18GPT 5.2temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.67
19Claude Sonnet 4.6temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.648
20Claude Sonnet 3.7temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.643
21Gemini 2.5 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.641
22DeepSeek V4 Protemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.64
23GPT 4otemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.638
24Mercury 2.5temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.638
25GLM 5.3temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.635
26GPT 5temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.632
27GLM 5.3 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.628
28Claude Sonnet 5temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.624
29Gemini 2.0 Flash Litetemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.604
30Gemini 2.0 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.603
31Claude Opus 4.5temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.602
32Gemini 2.5 Flash-Litetemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.594
33Grok 4.3temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.581
34GPT OSS 120Btemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.574
35Llama 4 Scouttemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.574
36Gemma 4 26B A4Btemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.568
37GLM 5.2temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.566
38Gemma 4 31B ITtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.563
39MiMo V2 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.558
40DeepSeek V3temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.551
41GPT-5 minitemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.542
42GPT-4o minitemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.541
43Kimi K2 Instructtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.538
44Claude Sonnet 4temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.528
45GPT-5 nanotemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.525
46Ling 3.0 Flashtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.522
47Gemma 3 27B ITtemperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.518
48GLM 4.6temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.516
49Jev 1.13.0temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.516
50Claude Sonnet 4.5temperature: 1.0, top_p: 0.95· UzLiB Uzbek linguistic benchmark data docs0.508

Showing 1–50 of 74 results.

Data credits: UzLiB, Tahrirchi (Abror Shopulatov). Public LEADERBOARD.md numbers; repository MIT covers software and documentation, no separate results licence stated and none asserted.