-
Notifications
You must be signed in to change notification settings - Fork 18
Pull requests: meta-pytorch/autoparallel
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Support dynamic batch sizes via symbolic shapes
CLA Signed
This label is managed by the Meta Open Source bot.
#418
opened Apr 13, 2026 by
fmassa
Contributor
Loading…
Deprecate compile, enable_ac, and numerics_logger from AutoParallel API
CLA Signed
This label is managed by the Meta Open Source bot.
Pre-overlap collective bucketing pass for FSDP/DDP
CLA Signed
This label is managed by the Meta Open Source bot.
#412
opened Apr 3, 2026 by
fmassa
Contributor
Loading…
Support imbalanced routing MoE + inductor
CLA Signed
This label is managed by the Meta Open Source bot.
Insert contiguous clone before collectives that require it
CLA Signed
This label is managed by the Meta Open Source bot.
Remove upstreamed sharding rule overrides for constant_pad_nd, native_layer_norm, and native_layer_norm_backward
CLA Signed
This label is managed by the Meta Open Source bot.
Bump black from 22.3.0 to 26.3.1
CLA Signed
This label is managed by the Meta Open Source bot.
dependencies
Pull requests that update a dependency file
python
Pull requests that update python code
#363
opened Mar 12, 2026 by
dependabot
bot
Loading…
Model communication-computation overlap in the sharding ILP
CLA Signed
This label is managed by the Meta Open Source bot.
Add factor-based and independent per-mesh-dim ILP sharding optimizers
CLA Signed
This label is managed by the Meta Open Source bot.
[overlap] Overlap simulation on 1d, 2d variants of llama3, DSv3 for 64, 256 gpus
CLA Signed
This label is managed by the Meta Open Source bot.
#318
opened Feb 18, 2026 by
IvanKobzarev
Contributor
Loading…
Add support for arbitrary inputs and outputs in AutoParallel
CLA Signed
This label is managed by the Meta Open Source bot.
#311
opened Jan 30, 2026 by
fmassa
Contributor
Loading…
[wip] update di/dw partition pass to use new api
CLA Signed
This label is managed by the Meta Open Source bot.
Use upstream PyTorch cost models instead of local copy
CLA Signed
This label is managed by the Meta Open Source bot.
[WIP][overlap] Overlap simulation on 1d, 2d variants of llama3 for 64, 256 gpus
CLA Signed
This label is managed by the Meta Open Source bot.
#289
opened Dec 19, 2025 by
IvanKobzarev
Contributor
•
Draft
Add support for non-tensor inputs in AutoParallel
CLA Signed
This label is managed by the Meta Open Source bot.
#264
opened Nov 21, 2025 by
fmassa
Contributor
Loading…
[WIP][Draft] PP + simplefsdp
CLA Signed
This label is managed by the Meta Open Source bot.
#253
opened Nov 17, 2025 by
IvanKobzarev
Contributor
•
Draft
Prototype to run AutoParallel PP with Local Tensor
CLA Signed
This label is managed by the Meta Open Source bot.
#252
opened Nov 14, 2025 by
dzmitry-huba
•
Draft
Add unit test for existing ac API behavior
CLA Signed
This label is managed by the Meta Open Source bot.
#244
opened Nov 11, 2025 by
ailzhang
Loading…
Estimate redistribute_cost by _TransformInfo
CLA Signed
This label is managed by the Meta Open Source bot.
#230
opened Nov 4, 2025 by
mori360
Loading…
Remove DTensor accumulateGrad overhead
CLA Signed
This label is managed by the Meta Open Source bot.
#193
opened Oct 7, 2025 by
fmassa
Contributor
Loading…
Enable reorder,bucketing,async_tp to example
CLA Signed
This label is managed by the Meta Open Source bot.
#174
opened Sep 30, 2025 by
IvanKobzarev
Contributor
Loading…
Previous Next
ProTip!
Add no:assignee to see everything that’s not assigned.