Benchmarks.bio - SpatialBench-Long: leaderboard
Long-form Benchmarks.bio spatial transcriptomics tasks that require multi-step biological data analysis, tool use, and synthesis over larger assay contexts.
Metric: Pass Rate (%). Source: benchmarks.bio. Status: years away from saturation. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.5 | 11.11 |
| 2 | Claude Opus 4.8 | 11.11 |
| 3 | Gemini 3.5 Flash | 11.11 |
| 4 | Claude Opus 4.6 | 9.72 |
| 5 | Claude Opus 4.7 | 8.33 |
| 6 | Grok 4.20 Beta (0309) (Reasoning) | 6.94 |
| 7 | GPT-5.4 | 5.56 |
| 8 | Kimi K2.6 | 5.56 |
| 9 | Claude Sonnet 4.6 | 4.17 |
| 10 | Gemini 3.1 Pro (Preview) | 4.17 |
| 11 | Grok 4.3 | 4.17 |
| 12 | Gemini 2.5 Pro | 1.39 |
Interactive version: theaggregate.ai/benchmark?slug=benchmarks-bio-spatialbench-long · How It Works · Data refreshed daily, snapshot 2026-09-05.