Researchbenchmarksscrna seqagents
scBench Evaluates Agents On scRNA-Seq Analysis
|
8.1
The authors introduce scBench, a benchmark of 394 verifiable problems derived from practical single-cell RNA sequencing (scRNA-seq) workflows, submitted Feb 9, 2026. It evaluates eight frontier AI models across six sequencing platforms and seven task categories, finding model accuracy of 29–53% and platform-dependent drops up to 40 percentage points. scBench complements SpatialBench and aims to measure and diagnose agent performance on real scRNA-seq data.
Scoring Rationale
Comprehensive benchmark addressing agent performance in real scRNA-seq, limited by preprint status and single-source evaluation.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
