Qwen3 30B A3B Instruct 8-bit: benchmark results

Provider: Alibaba.

Aggregate Rating 1596 ± 27, rank #1051 of 10916 rated models, from 28 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
Medmarks - LongHealth Task 288.63Score (%)72.2
Medmarks - PubMedQA76.4Score (%)64.8
Medmarks - SuperGPQA Medicine Hard41.63Score (%)60.2
Medmarks - MedQA86.83Score (%)58
Medmarks - MMLU-Pro Health74.98Score (%)56.8
Medmarks - MedCalc-Bench56.82Score (%)56.8
Medmarks - SuperGPQA Medicine Easy55.74Score (%)56.8
Medmarks - M-ARC42.67Score (%)56.2
Medmarks - CareQA EN88.67Score (%)54.5
Medmarks - MedXpertQA Understanding25.18Score (%)54.5
Medmarks - MedMCQA71.48Score (%)53.4
Medmarks - MedConceptsQA Easy92.08Score (%)48.9

Interactive version: theaggregate.ai/model?slug=qwen3-30b-a3b-instruct-8-bit · How It Works · Data refreshed daily, snapshot 2026-11-08.