Latch Bio
VariantBench: Genetic Variant Discovery and Interpretation
Agentic genetic variant analysis, from raw sequencing data to variant interpretation.
genomicsvariant-analysiscodetabularsequence
Status
curated
Saturation
unknown
Reproduction
not yet
Published results
38 results
Editor's note
Official current V1 results cover 118 evaluation records. Model-and-harness configurations are retained separately.
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 | 52.8% macro-mean pass rate | V1 · Claude Code · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5.5 | 51.7% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5.5 | 51.1% macro-mean pass rate | V1 · Claude Code · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5 | 50.6% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5.5 | 50.6% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5.5 | 49.7% macro-mean pass rate | V1 · Claude Code · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5.5* | 48.9% macro-mean pass rate | V1 · Claude Code · LSVP · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 5.5* | 48.6% macro-mean pass rate | V1 · Pi · LSVP · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Astra | 46.6% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6.1 Sol | 46.1% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Grok 4.7 | 45.5% macro-mean pass rate | V1 · Grok Build · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Astra | 45.2% macro-mean pass rate | V1 · Codex · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6.1 Sol | 43.2% macro-mean pass rate | V1 · Codex · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.8 | 42.4% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.8 | 42.4% macro-mean pass rate | V1 · Claude Code · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Fable 5.1* | 40.7% macro-mean pass rate | V1 · Claude Code · LSVP · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Grok 4.6 | 40.4% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5 | 39.8% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Grok 4.6 | 39.0% macro-mean pass rate | V1 · Grok Build · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 5 | 39.0% macro-mean pass rate | V1 · Claude Code · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Sol | 38.7% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.6 Sol | 35.3% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.7 | 35.0% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.6 Sol | 34.2% macro-mean pass rate | V1 · Codex · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Opus 4.7 | 33.0% macro-mean pass rate | V1 · Claude Code · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Gemini 3.7 Flash | 31.4% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Gemini 3.8 Flash | 31.4% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Kimi K3 | 31.4% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| DeepSeek V4.1 Flash | 31.1% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 4.6 | 31.1% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.5 | 30.5% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-6 Luna | 27.7% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.6 Luna | 27.1% macro-mean pass rate | V1 · Codex · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Claude Sonnet 4.6 | 26.3% macro-mean pass rate | V1 · Claude Code · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| GPT-5.6 Luna | 24.0% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Mistral Large 4.0 Preview | 22.0% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Nemotron 3 Ultra 550B A55B | 17.8% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |
| Nemotron 3.5 Lightning | 14.4% macro-mean pass rate | V1 · Pi · 118 evaluation records | Latch Bio · V1 leaderboard 2026-10-07 |