PhageBench - Host Prediction: leaderboard
Metric: Accuracy (%) on 1,200 four-option items naming the bacterial host order, family or genus, each a raw-nucleotide multiple-choice question with shuffled options, zero-shot chain-of-thought prompting (reasoning models at reasoning effort medium with at most 2,048 reasoning tokens); higher is better. Source: arxiv.org. Saturation forecast: Around November 2027. 8 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash (Medium) | 62.5 |
| 2 | Claude Sonnet 4.5 (Thinking) | 44.17 |
| 3 | GPT-OSS-120B | 41 |
| 4 | GPT-5.2 (Medium) | 35.25 |
| 5 | Qwen 3 Max (Thinking) | 34.75 |
| 6 | GPT-4o Mini | 27.75 |
Interactive version: theaggregate.ai/benchmark?slug=phagebench-host-prediction · How It Works · Data refreshed daily, snapshot 2026-10-07.