benchmarks.bio is hosted on AWS. It measures traffic with FullStory and Vercel Web Analytics.
| Technology | Version | First seen |
|---|---|---|
| Frontend Build Tools & CSS3 | ||
| – | 9 Nov 2026 | |
| – | 9 Nov 2026 | |
| – | 9 Nov 2026 | |
| CDN & Hosting2 | ||
| – | 9 Nov 2026 | |
| Analytics & Tag Management2 | ||
| – | 9 Nov 2026 | |
| Accessibility1 | ||
| – | 9 Nov 2026 | |
| Error Tracking & Performance Monitoring1 | ||
| – | 9 Nov 2026 | |
| Fonts & Icon Fonts3 | ||
| – | 9 Nov 2026 | |
Registration from the registry's RDAP record, the certificate the site serves, and its rank in public traffic lists.
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
Counted from every stack change we've recorded: of the sites we watched running one of this site's tools, how many went on to add another. A pattern across sites, not a forecast for this one.
dns4.p09.nsone.netdns1.p09.nsone.netdns3.p09.nsone.netdns2.p09.nsone.netUser-agent: OAI-SearchBot Allow: / User-agent: ChatGPT-User Allow: / User-agent: * Allow: / Sitemap: https://benchmarks.bio/sitemap.xml
# benchmarks.bio > AI agent leaderboards on real experimental biological data, reported per model and harness. Current version: V1. Data updated: 2026-11-08. V0 is historical; different evaluation sets and aggregation rules mean scores are not directly comparable. ## Current V1 results - [Overall leaderboard](https://benchmarks.bio/results/v1/) — readable HTML, no JavaScript required. - [Interactive leaderboard](https://benchmarks.bio/) — filterable charts and benchmark comparisons. - [JSON](https://benchmarks.bio/results/v1/latest.json) — complete structured results and provenance. - [CSV](https://benchmarks.bio/results/v1/latest.csv) — capabilities rankings. - [Markdown](https://benchmarks.bio/results/v1/index.md) — both leaderboards and methodology. ## Benchmarks - [SpatialBench](https://benchmarks.bio/spatial) — AI agents on spatial transcriptomics; V1, 115 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/spatialbench/index.md) · [Paper PDF](https://benchmarks.bio/papers/spatialbench.pdf) - [SpatialBench-Long](https://benchmarks.bio/spatial-long) — long-horizon spatial biology agents; V1, 21 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/spatialbench-long/index.md) · [Paper PDF](https://benchmarks.bio/papers/spatialbench-long.pdf) - [scBench](https://benchmarks.bio/scbench) — AI agents on single-cell RNA-seq; V1, 195 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/scbench/index.md) · [Paper PDF](https://benchmarks.bio/papers/scbench.pdf) - [scBench-Long](https://benchmarks.bio/scbench-long) — long-horizon single-cell analysis agents; V1, 22 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/scbench-long/index.md) · [Paper PDF](https://benchmarks.bio/papers/scbench-long.pdf) - [EpiBench](https://benchmarks.bio/epibench) — AI agents on epigenomics; V1, 106 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/epibench/index.md) · [Paper PDF](https://benchmarks.bio/papers/epibench.pdf) - [VariantBench](https://benchmarks.bio/variant) — AI agents on genetic variant analysis; V1, 118 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/variantbench/index.md) · [Paper PDF](https://benchmarks.bio/papers/variantbench.pdf) - [MetagenomicsBench](https://benchmarks.bio/metagenomics) — AI agents on metagenomics; V1, 100 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/metagenomics/index.md) · [Paper PDF](https://benchmarks.bio/papers/metagenomics.pdf) - [TxBench-Antibody-Discovery](https://benchmarks.bio/txbench-ab) — AI agents on antibody discovery; V1, 100 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/txbench-ab/index.md) · [Paper PDF](https://benchmarks.bio/papers/txbench-ab.pdf) - [TxBench-Antibody-Development](https://benchmarks.bio/txbench-abdev) — AI agents on antibody development; V1, 60 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/txbench-abdev/index.md) · [Paper PDF](https://benchmarks.bio/papers/txbench-ab-dev.pdf) - [TxBench-Preclinical-Pharmacology](https://benchmarks.bio/txbench) — AI agents on preclinical pharmacology; V1, 100 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/txbench-pp/index.md) · [Paper PDF](https://benchmarks.bio/papers/txbench-pp.pdf) - [TxBench-Oligo-Discovery](https://benchmarks.bio/txbench-oligo) — AI agents on oligonucleotide discovery; V1, 113 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/txbench-od/index.md) · [Paper PDF](https://benchmarks.bio/papers/txbench-oligo.pdf) - [NeuroscienceBench](https://benchmarks.bio/neuroscience) — AI agents on neuroscience research data; V1, 91 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/neurosciencebench/index.md) - [RareDiseaseBench](https://benchmarks.bio/raredisease) — AI agents on rare disease diagnosis; V1, 35 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/rarediseasebench/index.md) · [Paper PDF](https://benchmarks.bio/papers/rarediseasebench.pdf) - [BioSecBench-Surveillance](https://benchmarks.bio/surveillance) — AI agents on biosecurity surveillance; V1, 102 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/biosecbench-surveillance/index.md) · [Paper PDF](https://benchmarks.bio/papers/biosecbench-surveillance.pdf) - [BioSecBench-RespiratoryVirus-Surveillance](https://benchmarks.bio/respvirus-surveillance) — AI agents on respiratory virus surveillance; V1, 110 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/biosecbench-respvirus-surveillance/index.md) · [Paper PDF](https://benchmarks.bio/papers/respvirus-surveillance.pdf) - [BioSecBench-Refusal](https://benchmarks.bio/security) — AI biology safeguards and refusals; V1, 107 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/biosecbench-refusal/index.md) · [Paper PDF](https://benchmarks.bio/papers/biosecbench-refusal.pdf) - [BioSecBench-Function](https://benchmarks.bio/function) — AI agents on biological function analysis; V1, 111 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/functionbench/index.md) · [Paper PDF](https://benchmarks.bio/papers/functionbench.pdf) - [OncologyBench](https://benchmarks.bio/oncology) — AI agents in oncology; V1, 25 evaluations; updated 2026-11-08. [Markdown](https://benchmarks.bio/benchmarks/oncologybench/index.md) ## Interpretation and sources - LSVP: Anthropic's safety-filter-exempt evaluation variant, which produces fewer refusals on biology prompts. LSVP results are reported alongside the model's standard run and are not a replacement for it. - Overall capability scores weight observed benchmarks by configured evaluation counts and allow at most two missing benchmarks. Missing scores are excluded, not zero. Safeguards is scored separately. - [Methodology and provenance](https://benchmarks.bio/results/v1/#provenance) — full scoring and import notes. - [Version guide](https://benchmarks.bio/results/) · [Version manifest](https://benchmarks.bio/results/versions.json). ## Historical V0 - [V0 leaderboard](https://benchmarks.bio/results/v0/) · [JSON](https://benchmarks.bio/results/v0/latest.json) · [CSV](https://benchmarks.bio/results/v0/latest.csv) · [Markdown](https://benchmarks.bio/results/v0/index.md).
| – |
| 9 Nov 2026 |
| – |
| 9 Nov 2026 |
| – |
| 9 Nov 2026 |
| – | 9 Nov 2026 |