Skip to content

Kazakh

Қазақ тілі
Highest accuracy
Solar Pro 4Upstage · High reasoning
96.0%

4.0 points ahead of the next model

A close alternative
Mercury 2.5 PreviewInception · High reasoning
92.0%

4.0 points behind the highest score

How much scores vary
Lowest score32.0%Highest score96.0%
64.0pp

Between the highest and lowest of 10 published results.

How the models compare

Published results, ranked by accuracy. Higher is better.

Compare side by side
Kazakh · How the models compare
#ModelAccuracyGap to top
1Solar Pro 4High reasoning
Solar Pro 4: 96.0%96.0%
Top score
2Mercury 2.5 PreviewHigh reasoning
Mercury 2.5 Preview: 92.0%92.0%
-4.0 pp
2Qwen3.7 FlashHigh reasoning
Qwen3.7 Flash: 92.0%92.0%
-4.0 pp
3gpt-oss-120bHigh reasoning
gpt-oss-120b: 88.0%88.0%
-8.0 pp
3Solar Pro 4Low reasoning
Solar Pro 4: 88.0%88.0%
-8.0 pp
4Qwen3.7 FlashLow reasoning
Qwen3.7 Flash: 84.0%84.0%
-12.0 pp
5gpt-oss-120bLow reasoning
gpt-oss-120b: 80.0%80.0%
-16.0 pp
6Mercury 2.5 PreviewLow reasoning
Mercury 2.5 Preview: 64.0%64.0%
-32.0 pp
7Ling 3.0 FlashLow reasoning
Ling 3.0 Flash: 36.0%36.0%
-60.0 pp
8Ling 3.0 FlashHigh reasoning
Ling 3.0 Flash: 32.0%32.0%
-64.0 pp

The numbers, with the working attached.

Tests, prompts, settings and model responses live on the release pages.

View published releases