Item detail

AtomFlow-AI/MoleCode

AtomFlow-AI/MoleCode is an MIT-licensed LLM-native graph-explicit molecular language that makes a molecule's structure the language itself — atoms are typed `prefix_Element_Number[Label]` nodes, bonds are `---` / `===` / `-.-` edges with `===|E|` / `===|Z|` and `_R` / `_S` for stereochemistry, serialized as a Mermaid graph that is deterministically and losslessly inter-convertible with SMILES / MO

Score8.5
Popularity286.0
Risklow
TierGold
Score breakdown
Usefulness8.9
Novelty10.0
Momentum10.0
Maturity9.1
Open-source/build7.4
Evidence7.2
Workflow potential9.2
Setup ease6.5

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for cheminformatics / computational-chemistry / drug-discovery teams that need an LLM-friendly representation for molecules that does not require the model to reconstruct graph structure from SMILES strings (the dominant accuracy bottleneck for SMILES-based LLM chemistry); for AI-chemistry researchers who want a single language that covers small molecules, polymers (with `×n` repeat units a

Who should use it

BuildersPower users

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

Risk label needs manual review.

Evidence links

Closest alternatives / related signals