Latch Bio
TxBench: Preclinical Pharmacology
Interpretation of laboratory and animal data for small-molecule drugs before human trials.
therapeuticspharmacologycodetabulartext
Status
curated
Saturation
unknown
Reproduction
not yet
Published results
37 results
Editor's note
Official current V1 results cover 100 evaluation records, including target engagement, safety and disease-model evidence.
Published results
V1 · current official leaderboard
Current V1 (formerly labeled V2) uses updated evaluation sets and scoring; separate from legacy V0 and paper results. A macro-mean is not a solved-task count. Starred models use Anthropic's LSVP and remain distinct from standard configurations.
| Model | Score | Variant | Source |
|---|---|---|---|
| Claude Opus 5.5* | 65.3% macro-mean pass rate | V1 · Pi · LSVP · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5.5* | 65.0% macro-mean pass rate | V1 · Claude Code · LSVP · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5.5 | 64.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5 | 63.0% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Sol | 63.0% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5.5 | 62.3% macro-mean pass rate | V1 · Claude Code · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Grok 4.6 | 61.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.6 Sol | 61.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Fable 5.1* | 59.7% macro-mean pass rate | V1 · Claude Code · LSVP · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6.1 Sol | 59.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5 | 59.7% macro-mean pass rate | V1 · Claude Code · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Astra | 59.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Astra | 59.3% macro-mean pass rate | V1 · Codex · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.8 | 57.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Grok 4.6 | 57.0% macro-mean pass rate | V1 · Grok Build · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.5 | 55.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Gemini 3.7 Flash | 55.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6.1 Sol | 55.0% macro-mean pass rate | V1 · Codex · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Grok 4.7 | 54.7% macro-mean pass rate | V1 · Grok Build · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Gemini 3.8 Flash | 54.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.8 | 54.3% macro-mean pass rate | V1 · Claude Code · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5 | 49.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5 | 49.3% macro-mean pass rate | V1 · Claude Code · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| DeepSeek V4.1 Flash | 46.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.7 | 45.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Luna | 44.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.6 Luna | 43.7% macro-mean pass rate | V1 · Codex · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.6 Luna | 43.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.7 | 42.3% macro-mean pass rate | V1 · Claude Code · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Kimi K3 | 39.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 4.6 | 34.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 4.6 | 33.3% macro-mean pass rate | V1 · Claude Code · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5.5 | 30.7% macro-mean pass rate | V1 · Claude Code · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Mistral Large 4.0 Preview | 29.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5.5 | 26.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Nemotron 3 Ultra 550B A55B | 22.7% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Nemotron 3.5 Lightning | 10.3% macro-mean pass rate | V1 · Pi · 100 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |