Skip to content

Pull requests: pytorch/torchtitan

New pull request New
73 Open 849 Closed
73 Open 849 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[WIP][kernels][blackwell] Cutlass blackwell grouped gemm using cute dsl CLA Signed This label is managed by the Meta Open Source bot.
#1276 opened Jun 8, 2025 by lessw2020 Loading…
run core tests only if core files are modified CLA Signed This label is managed by the Meta Open Source bot.
#1275 opened Jun 8, 2025 by tianyu-l Loading…
[deepseek][blackwell] add Cutlass cute dsl blackwell dense based looping group gemm CLA Signed This label is managed by the Meta Open Source bot.
#1274 opened Jun 8, 2025 by lessw2020 Loading…
[SimpleFSDP] Add support for SimpleFSDP DCP CLA Signed This label is managed by the Meta Open Source bot.
#1273 opened Jun 8, 2025 by ruisizhang123 Loading…
3
[deepseek][blackwell] add manual looping group gemm to enable base working inference on Blackwell CLA Signed This label is managed by the Meta Open Source bot.
#1272 opened Jun 7, 2025 by lessw2020 Loading…
add logging to keep track of internal usage CLA Signed This label is managed by the Meta Open Source bot.
#1271 opened Jun 6, 2025 by tianyu-l Loading…
set up action to publish to PyPI on release CLA Signed This label is managed by the Meta Open Source bot.
#1270 opened Jun 6, 2025 by tianyu-l Loading…
[llama4] enable expert parallel on the same device mesh as tp (tp2ep) CLA Signed This label is managed by the Meta Open Source bot.
#1269 opened Jun 6, 2025 by hann-wang Loading…
1
Fix lr scheduler CLA Signed This label is managed by the Meta Open Source bot. release blocking Issues that are blocking the milestone / release completion
#1261 opened Jun 4, 2025 by CarlosGomes98 Loading… torchtitan v0.1.0 release
5
Added support for creating ROCm docker image for torchtian & run torchtitan tests on ROCm. CLA Signed This label is managed by the Meta Open Source bot. module: rocm
#1260 opened Jun 4, 2025 by akashveramd • Draft
2
[WIP][Blackwell Kernels] Blackwell group gemm and dense gemms with Python Cutlass CLA Signed This label is managed by the Meta Open Source bot.
#1256 opened Jun 3, 2025 by lessw2020 Loading…
alternative implementation of create_indices_from_offsets_nosync compatible with torch.compile CLA Signed This label is managed by the Meta Open Source bot.
#1251 opened Jun 1, 2025 by hann-wang Loading…
2
[float8] add float8 rowwise MoE prototype CLA Signed This label is managed by the Meta Open Source bot.
#1245 opened May 30, 2025 by danielvegamyhre • Draft
3
Add AMD GPU node for integration test CLA Signed This label is managed by the Meta Open Source bot.
#1241 opened May 29, 2025 by mori360 • Draft
3
Implement initial_load_path for checkpointer CLA Signed This label is managed by the Meta Open Source bot. release blocking Issues that are blocking the milestone / release completion
#1236 opened May 28, 2025 by fegin Loading… torchtitan v0.1.0 release
12
[cp][flex_attention] integration test trial CLA Signed This label is managed by the Meta Open Source bot.
#1228 opened May 27, 2025 by XilunWu • Draft
[Flux] Add batched inference CLA Signed This label is managed by the Meta Open Source bot.
#1227 opened May 27, 2025 by CarlosGomes98 Loading…
32
[WIP] Implement the feature to save unsharded weights at the last step CLA Signed This label is managed by the Meta Open Source bot.
#1219 opened May 23, 2025 by fegin Loading…
4
[WIP][Experimental] Activation Offloading CLA Signed This label is managed by the Meta Open Source bot.
#1218 opened May 23, 2025 by lessw2020 Loading…
2
[WIP][DeepSeek] DeepSeek training and component integration with Titan main components CLA Signed This label is managed by the Meta Open Source bot.
#1183 opened May 13, 2025 by lessw2020 Loading…
3
compile: turn off fullgraph=True to support llama4 CLA Signed This label is managed by the Meta Open Source bot.
#1182 opened May 12, 2025 by bdhirsh Loading…
15
🐛 Use correct path for train_configs
#1163 opened May 2, 2025 by brianlechthaler Loading…
2
[cp][flex_attention] integration test trial CLA Signed This label is managed by the Meta Open Source bot. module: context parallel
#1160 opened May 1, 2025 by XilunWu • Draft
5
[WIP] float8 rowwise all gather CLA Signed This label is managed by the Meta Open Source bot.
#1157 opened Apr 30, 2025 by danielvegamyhre • Draft
[WIP] token-expert assignments and layer affinity tracking for expert placement via ILP solving CLA Signed This label is managed by the Meta Open Source bot.
#1152 opened Apr 28, 2025 by lessw2020 Loading…
ProTip! Adding no:label will show everything without a label.