omtx/lula-1
Other • Updated • 197 • 2
AI & ML interests
foundation models for small molecule discovery
Recent Activity
Organization Card
+
+
+
omtx.ai
Decoding the language of small molecule discovery
Open-weight protein–ligand scoring for small molecule discovery.
LULA-1 is a sequence-only protein–ligand binding score: protein sequence plus ligand SMILES in, binding score out. No structure, no docking, no folding step — built for local inference and fine-tuning.
Sequence-only Local inference Fine-tuning
01
LULA-1 local
Open-weight LULA-1 for local inference and fine-tuning, with runnable examples and checksum verification.
02
Hosted Om scorer
Hosted LULA workflows for teams that want managed runs, tracking, and scale behind one API.
03
Molecule Fulfillment
Score candidates, select molecules, and request quotes to move prioritized compounds toward synthesis.
+
Release status
Open-weight LULA-1 is being prepared for public release.
The launch package covers the model card, license, manifest, checksums, local scoring examples, and a smoke dataset. Links appear here as each item goes live.
Weights — preparing Model card — preparing License — preparing Checksums — preparing Examples — preparing datasets 0
None public yet