Skip to content

Pull requests: Lightning-AI/lightning-thunder

New pull request New
28 Open 1,027 Closed
28 Open 1,027 Closed
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
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[WIP] Add utility function to benchmark performance of fusion region with nvfuser and torch.compile
#1682 opened Jan 22, 2025 by kshitij12345 • Draft
2
make no_sync behave like in pytorch
#1681 opened Jan 22, 2025 by t-vi • Draft
1
add debug option to show interpreter progress
#1680 opened Jan 22, 2025 by t-vi Loading…
fixing dce pass to remove duplicate shape queries.
#1673 opened Jan 21, 2025 by jjsjann123 Loading…
1
[WIP] Add cross_entropy to torchcompile_cat executor
#1655 opened Jan 17, 2025 by riccardofelluga • Draft
1
2
[WIP] initial changes to accommodate pipeline parallelism transform
#1650 opened Jan 16, 2025 by ali-alshaar7 • Draft
4 tasks
[WIP] Thunderfx report
#1636 opened Jan 10, 2025 by kiya00 • Draft
4 tasks
3
revise examine memory's collection handling, output memory annotation in benchmark_litgpt
#1634 opened Jan 10, 2025 by t-vi • Draft
Remove requires_grad from regular TensorProxy and add specialized TensorProxy with requires_grad and is_leaf attributes developer efficiency
#1599 opened Dec 30, 2024 by IvanYashchuk • Draft
7
Allow Proxy creation without a set TraceCtx tracing architecture
#1598 opened Dec 30, 2024 by IvanYashchuk • Draft
6
[tensor subclass] print type_string of tensor attributes
#1592 opened Dec 29, 2024 by crcrpar Loading…
1
check tensor attrs of tensor wrapper subclasses in prologue
#1591 opened Dec 28, 2024 by crcrpar Loading…
[tensor wrapper subclass] Add support for torchao.float8 mlp
#1585 opened Dec 23, 2024 by crcrpar • Draft
1
[tensor wrapper subclass] Add trace transform for tensor subclasses documentation Improvements or additions to documentation
#1584 opened Dec 23, 2024 by crcrpar • Draft
[tensor wrapper subclass] Add Proxy, prim, and lookaside
#1583 opened Dec 23, 2024 by crcrpar • Draft
8
[nvfuser executor] Allow sm89 for fp8 types
#1576 opened Dec 20, 2024 by crcrpar • Draft
6
[benchmark] add option to enable CompiledAutograd
#1536 opened Dec 10, 2024 by crcrpar • Draft
Refactor autocast to AutocastTransform for Improved Composability and Transform Integration
#1516 opened Dec 4, 2024 by rittik9 • Draft
1 of 4 tasks
1
2
Fix TransformerEngine and Activation checkpointing interaction for thunderFX
#1473 opened Nov 26, 2024 by kshitij12345 • Draft
2
1
update ctxs of bsym representing torch.autograd.Funcion
#1462 opened Nov 21, 2024 by crcrpar • Draft
[torchao float8tensor] documentation Improvements or additions to documentation
#1415 opened Nov 8, 2024 by crcrpar • Draft
24
Rework timer infrastructure
#1410 opened Nov 7, 2024 by tfogal Loading…
10
[Tensor Subclasses] Trace transfom to interpret __torch_dispatch__
#1394 opened Nov 3, 2024 by crcrpar • Draft
31
Allow linear to be consumed by nvFuser by default memory use nvfuser
#1371 opened Oct 30, 2024 by IvanYashchuk Loading…
21
Add an option to print nvFuser repros.
#1362 opened Oct 29, 2024 by tfogal Loading…
14
ProTip! What’s not been updated in a month: updated:<2024-12-23.