Benchmark RadarAI BENCHMARK PROFILE
TherapeuticsBench
TxBench-PP evaluates AI agents on small-molecule preclinical pharmacology tasks including mechanism-of-action, pharmacodynamics, and safety reasoning, using realistic workflow snapshots and deterministic scoring.
- Released
- 2026-06-17
- Readiness
- Paper only
- Primary field
- Science & Research
Why it matters
Provides verifiable evaluation of agents on realistic drug discovery decisions, addressing the need for trusted benchmarks in high-stakes scientific applications.
Motivation
Artificial intelligence (AI) agents promise to accelerate drug discovery by compressing interpretation and decision-making loops, but practical deployment requires trusted evaluation on realistic program decisions.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.