Skip to content

Pull requests: aqlaboratory/openfold-3

Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort

Pull requests list

Chunk large LayerNorm inputs to work around torch bug
#231 opened May 26, 2026 by GMNGeoffrey Contributor Loading…
Make config chunk_size maximum rather than minimum safe-to-test Internal only label used to indicate PRs that are ready for automated CI testing.
#227 opened May 20, 2026 by GMNGeoffrey Contributor Loading…
Treat arg-cache length mismatch as a cache miss in ChunkSizeTuner
#224 opened May 18, 2026 by GMNGeoffrey Contributor Loading…
Install py-lmdb from conda + unpin python + create openfold3-msa safe-to-test Internal only label used to indicate PRs that are ready for automated CI testing.
#222 opened May 17, 2026 by sdvillal Contributor Loading…
Avoid flattening and expanding broadcast dimensions when chunking
#213 opened May 5, 2026 by GMNGeoffrey Contributor Loading…
2
2
Replace scipy's truncnorm.rvs with nn.init.trunc_normal_ to make model load faster safe-to-test Internal only label used to indicate PRs that are ready for automated CI testing.
#178 opened Apr 13, 2026 by rizsp Contributor Loading…
use torch.gather() to make batched_gather() more efficient enhancement New feature or request model Related to the model definition sketch This is an idea/sketch rather than a mergeable PR, significant work needed to advance needed training Relating to the training pipeline
#135 opened Mar 14, 2026 by OccupyMars2029 Draft
Fix check_sequence filter
#80 opened Dec 28, 2025 by gnikolenyi Collaborator Loading…
Preserve Raw ColabFold MSA Output Files
#35 opened Nov 12, 2025 by qurat-ul-ain95 Contributor Loading…
ProTip! Updated in the last three days: updated:>2026-05-25.