LIBRA - ruQasper — leaderboard

Metric: Dataset Total Score (%). Source: huggingface.co. 17 models tracked.

Top models

#ModelScore
1GPT-4o31.72
2Mistral Nemo24.47
3Llama 3.1 8B6.52
4Mistral-7B-v0.35.75
5Mistral 7B5.38
6GLM-4 9B Chat5.01
7LongChat-7B-v1.5-32k4.97
8Llama 2 7B4.68
9T-lite-instruct-0.12.73
10ChatGLM2 6B2.58
11Mistral-7B-v0.12.48
12Llama 3 8B2.18

Interactive version: theaggregate.ai/benchmark?slug=libra-ruqasper · How the rankings work · Data refreshed daily, snapshot 2026-07-22.