An open drug-discovery challenge you can enter with an LLM
Early-stage drug discovery is usually closed off - not because the questions are unapproachable, but because the tooling is. The Open Discovery Challenge is an attempt to remove that gate: propose a molecule, get it scored on a public rubric, and see where it lands.
Two seasons are running right now
Season 1 - Malaria. Target: PfDHODH in Plasmodium falciparum. Counter-target: human DHODH. Closes 30 Sep 2026. Prize $1,000.
Season 2 - Tuberculosis. Target: InhA in Mycobacterium tuberculosis. Counter-target: human FASN (ER domain). Closes 31 Oct 2026. Prize $2,000.
How you enter
Submit a molecule as SMILES or InChI. Use Claude, GPT, Gemini, Qwen, KIMI, DeepSeek, or your own model - the challenge does not care how you arrived at the structure, only what the structure is.
How it is scored
100 points across six axes: activity, binding, selectivity, ADMET, novelty and synthesizability.
The weights differ by season. Season 1 leans on activity (30 points). Season 2 splits weight toward binding and selectivity (25 each), because an InhA inhibitor that also hits the human enzyme is not useful.
The counter-target is the interesting part
It is easy to score well on "binds the pathogen enzyme". It is much harder to bind the pathogen enzyme and leave the human homolog alone. That is why every season carries a counter-target, and why selectivity is weighted as heavily as binding.
You can keep your structure private
The leaderboard shows a masked SMILES for private submissions, so you can compete without publishing the molecule.
The leaderboard is an API
/api/seasons and /api/leaderboard?season=N return JSON, so you can pull standings, weights and per-axis breakdowns programmatically.
One honest note
A high score is a research hypothesis, not a drug. Every number here comes from computational scoring. It is not clinical validation, and it does not become one by ranking well.
Challenge: https://huggingface.co/spaces/FINAL-Bench/open-discovery-challenge
Top comments (0)