SpatialBench
A benchmark of deterministic, verifiable agentic problems derived from real spatial-transcriptomics workflows, testing whether agents can manipulate data and recover key biological results.
Scientific task · Omics analysis
Analyze genome-scale profiles
组学与细胞分析
omics-cellular-analysisCoverage
A benchmark of deterministic, verifiable agentic problems derived from real spatial-transcriptomics workflows, testing whether agents can manipulate data and recover key biological results.
A benchmark for evaluating LLM cell-type annotation across scRNA-seq and single-cell multiomics data.
An agentic bioinformatics benchmark of objective, expert-authored mysteries over anonymized real-world biological data, scored on final answers rather than prescribed analysis paths.
A containerized benchmark of long-horizon bioinformatics analysis over real published notebooks and associated data, with open-answer and multiple-choice evaluation modes.
A 100-task agent benchmark of objectively gradable computational-biology problems requiring multi-step reasoning, bespoke code, tools, and real-world external resources.
A research-level agent benchmark of 129 synthetic, multistage computational-biology analyses that require iterative QC, statistical modeling, diagnostics, and decision-relevant judgment.
A 13-task benchmark of single-cell data integration across simulated, scRNA-seq, and scATAC-seq settings, evaluated with 14 batch-removal and biological-conservation metrics.
Each row keeps its original unit and basis. Rows with different units or overlapping mappings are never added.
| Benchmark | Mapped task | Coverage | Count | Version | Evidence |
|---|---|---|---|---|---|
| SpatialBench root: spatialbench | Cell-state clustering official-taxonomy · high | explicitly-in-scope | 3 problems official category_results.json n_evals | repo-159-5042c4f as of 2026-06-10 | spatialbench-evidence-current-counts |
| Single-cell Omics Arena root: single-cell-omics-arena-soar | Cell-type annotation official-taxonomy · high | explicitly-in-scope | 1226 tasks Benchmark-wide cell-type annotation task total reported by the creator paper. | initial-release | single-cell-omics-arena-soar-automated-task-1-evidence |
| SpatialBench root: spatialbench | Cell-type annotation official-taxonomy · high | explicitly-in-scope | 45 problems official category_results.json n_evals | repo-159-5042c4f as of 2026-06-10 | spatialbench-evidence-current-counts |
| SpatialBench root: spatialbench | Differential expression analysis official-taxonomy · high | explicitly-in-scope | 40 problems official category_results.json n_evals | repo-159-5042c4f as of 2026-06-10 | spatialbench-evidence-current-counts |
| BioMysteryBench root: biomysterybench | Omics and cellular analysis official-taxonomy · high | explicitly-in-scope | Not reported v11 mystery-bioinformatics problems using omics and cellular data. | v11 | biomysterybench-evidence-taxonomy |
| BixBench root: bixbench | Omics and cellular analysis official-taxonomy · high | observed | Not reported v1.5 questions carrying official genomics, transcriptomics, epigenomics, single-cell, proteomics, or integrative-omics category labels. | v1.5 | bixbench-evidence-taxonomy |
| CompBioBench root: compbiobench | Omics and cellular analysis official-taxonomy · high | explicitly-in-scope | Not reported v1 tasks in official genomics, transcriptomics, epigenetics, single-cell, and spatial domains. | v1 | compbiobench-evidence-taxonomy |
| GeneBench-Pro root: genebench-pro | Omics and cellular analysis official-taxonomy · high | explicitly-in-scope | Not reported Problems assigned to transcriptomics, epigenomics, single-cell, spatial, proteomics, microbiome, and related official domains. | paper-v1 | genebench-pro-paper-evidence |
| scIB root: scib | Single-cell data integration official-taxonomy · high | explicitly-in-scope | 13 tasks Atlas-level integration tasks. | paper-2021 | scib-paper-definition-evidence |
| SpatialBench root: spatialbench | Spatial omics analysis official-taxonomy · high | explicitly-in-scope | 36 problems official category_results.json n_evals | repo-159-5042c4f as of 2026-06-10 | spatialbench-evidence-current-counts |
Runs are included only for benchmark records mapped here (and formal child tracks when a mapped suite is the root). A task mapping does not imply that every run isolates this task.