ToxReason - Cardiotoxicity: leaderboard

ToxReason (2026) tests mechanistic chemical toxicity reasoning: given a query molecule with experimental initiating-event evidence from structurally similar compounds, the model predicts liver, heart and kidney toxicity and explains how molecular initiating events lead to the adverse outcome along an Adverse Outcome Pathway. F1 for the heart toxicity label, in percent; higher is better.

Metric: F1 (%) for heart toxicity predicted by the model on the ToxReason test set (query chemicals with curated human CTD toxicity labels and adverse outcome pathway context, eight retrieved activation and inhibition examples from similar compounds), zero-shot, temperature 0; higher is better. Source: arxiv.org. Saturation forecast: Around July 2027. 11 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1DeepSeek R1 Distill Llama 70B78.5#1669
2Llama 3.1 70B Instruct76.1#1571
3GPT-4o74.6#751
4GPT-572.7#178
5O372.5#237
6O4 Mini71.3#368
7GPT-5.171.2#240
8Qwen 3 4B Instruct71.1#1696
9Gemma 3 27B (IT)69.1#1324
10Qwen 2.5 14B Instruct65#1775
11Llama 3.1 8B Instruct60.2#4600

No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=toxreason-cardiotoxicity · How It Works · Data refreshed daily, snapshot 2026-11-21.