TxBench Oligo Discovery V1
Science
TxBench-Oligo-Discovery V1: 113 deterministic evaluation tasks using experimental biological data. Published pass rates distinguish model and agent harness.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 55.5 |
| Claude Opus 5 | 48.1 |
| Grok 4.6 | 44.0 |
| Opus 4.8 | 39.9 |
| Sonnet 5 | 39.3 |
| Gemini 3.5 Flash | 38.6 |
| Grok 4.5 | 38.4 |
| GPT-5.6 Sol | 34.5 |
| DeepSeek V4.1 Flash | 33.9 |
| GPT-5.5 | 33.6 |
| GPT-5.6 Terra | 33.3 |
| GPT-5.6 Luna | 23.9 |