Benchmarks.bio - SpatialBench: leaderboard
LatchBio agentic benchmark on messy real-world spatial transcriptomics data, with models writing and running analysis workflows across assays, platforms, and task categories.
Metric: Pass Rate (%). Source: benchmarks.bio. Saturation forecast: Around March 2027. 22 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Grok 4.6 | 77.39 | #26 |
| 2 | DeepSeek V4.1 Flash | 75.36 | #36 |
| 3 | GPT-5.6 Sol | 74.49 | #19 |
| 4 | Claude Opus 5 | 71.01 | #12 |
| 5 | Grok 4.5 | 71.01 | #48 |
| 6 | GPT-6 Astra | 70.14 | #6 |
| 7 | GPT-5.5 | 69.57 | #31 |
| 8 | GPT-5.4 | 69.28 | #79 |
| 9 | Claude Sonnet 5 | 69.28 | #57 |
| 10 | Claude Mythos 5 | 69.2 | #2 |
| 11 | Claude Opus 4.8 | 68.99 | #37 |
| 12 | Gemini 3 Flash | 66.96 | #99 |
| 13 | Gemini 3.5 Flash | 66.96 | #60 |
| 14 | Claude Opus 4.7 | 64.64 | #49 |
| 15 | Claude Opus 4.6 | 63.48 | #63 |
No result here: #3 Claude Opus 5.5, #7 GPT-6.1 Sol, #8 Gemini 4 Argon.
Interactive version: theaggregate.ai/benchmark?slug=benchmarks-bio-spatialbench · How It Works · Data refreshed daily, snapshot 2026-10-20.