Qwen2-Math-7B-Instruct: benchmark results

Provider: Alibaba. Released 2024-08-08. Access: Open.

Unified ELO 1418 ± 48, rank #1765 of 2656 rated models, from 25 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Omni-MATH29.36Overall Accuracy (%)57.1
MERA - LCS12.4Accuracy (%)47.6
MERA - MathLogicQA43.13Accuracy (%)41.4
MERA - RWSD53.46Accuracy (%)37.8
MERA - ruHateSpeech62.64Accuracy (%)32.5
MERA - ruModAr46.1EM (%)29.3
MERA - ruMultiAr23.63EM (%)28.7
MERA - BPS84.8Accuracy (%)27.8
MERA - PARus63.2Accuracy (%)25.2
MERA - RCB47.72Accuracy (%)23.9
Wolfram LLM Benchmarking Project24.1Correct Functionality (%)22.1
MERA - ruTiE56.2Accuracy (%)22

Interactive version: theaggregate.ai/model?slug=qwen2-math-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-19.