Molence Bio open-sources Mol-1-constraint, lifting antibody-antigen cofolding success to 88.3%
sokrypton · x · 2026-11-12
Molence Bio released Mol-1-constraint with open code and weights for all researchers. Benchmarks vs the base Mol-1 and open-source cofolding models:
- Antibody-antigen: 75.0% → 88.3% success on FoldBench-AB (DockQ >0.23)
- Protein-ligand: 36.1% → 63.9% on Runs N' Poses' 0-20 similarity bin, 27.8 points above Mol-1
- Protein-peptide: 54.1% → 60.2% at DockQ >0.80 on MolBench-Peptide-26
The team frames the release as 'research should benefit everyone,' with a hosted demo available.
Related event: Molence Bio Open-Sources Mol-1-constraint for Structure Prediction(2 posts)→
More from Research
- Dejan's LLM brand study probes 193k brand mentions for just $391.71 using Gemini — dejanseo · 2026-11-13
- Galaxea Co-founder Hang Zhao's CoRL Keynote: Split Robot Intelligence into System 2/1/0 by Timescale — YuXiang_IRVL · 2026-11-13
- A Summary of the Macy Conferences: The Birthplace of Cybernetics and Early AI Thought — ArtificialOther · 2026-11-13
- vismatch: unified API wrapping 50+ image matching models, 943 stars on GitHub — gabriberton · 2026-11-13
- Stanford's Speculative Self-Distillation cuts training tokens 38-69% with less forgetting — khademinori · 2026-11-13
- EEPO: Adaptive Sample-Then-Forget Rollouts Break Diversity Collapse in RLVR Training — burkov · 2026-11-13