Skip to content

Pull requests: pytorch/torchtitan

New pull request New
87 Open 906 Closed
87 Open 906 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

[DSV3] Explicitly convert to bfloat16 when use grouped mm CLA Signed This label is managed by the Meta Open Source bot.
#1367 opened Jul 3, 2025 by wwwjn Loading…
4
[WIP] Compile for dp2ep CLA Signed This label is managed by the Meta Open Source bot.
#1365 opened Jul 3, 2025 by xmfan • Draft
non parallelized basic validator implementation [WIP] CLA Signed This label is managed by the Meta Open Source bot.
#1362 opened Jul 2, 2025 by wesleytruong Loading…
46
[benchmark] add h200 bench CLA Signed This label is managed by the Meta Open Source bot.
#1361 opened Jul 2, 2025 by asaiacai • Draft
4
Minimal reproduce of error when saving state dict after looping CLA Signed This label is managed by the Meta Open Source bot.
#1358 opened Jun 30, 2025 by wesleytruong Loading…
1
7
Add support for saving HF format tensors with DCP CLA Signed This label is managed by the Meta Open Source bot.
#1351 opened Jun 27, 2025 by ankitageorge • Draft
11
[WIP] Document MX FP8 recipe CLA Signed This label is managed by the Meta Open Source bot.
#1350 opened Jun 27, 2025 by lessw2020 Loading…
1
Autoparallel support for DP-only, DP+TP, or TP-only CLA Signed This label is managed by the Meta Open Source bot.
#1349 opened Jun 27, 2025 by wconstab Loading…
6
[WIP] Enable causal block mask for sdpa CLA Signed This label is managed by the Meta Open Source bot.
#1348 opened Jun 26, 2025 by mreso • Draft
3
[DSV3] Add PP support for DSV3 CLA Signed This label is managed by the Meta Open Source bot.
#1345 opened Jun 26, 2025 by H-Huang • Draft
8
[SimpleFSDP] Add support for hsdp+tp CLA Signed This label is managed by the Meta Open Source bot.
#1343 opened Jun 26, 2025 by ruisizhang123 Loading…
2
Refactor Tokenizer -> BaseTokenizer CLA Signed This label is managed by the Meta Open Source bot.
#1333 opened Jun 24, 2025 by H-Huang Loading…
13
[kernels][blackwell] add cutlass/cute group gemm forward for blackwell CLA Signed This label is managed by the Meta Open Source bot.
#1327 opened Jun 22, 2025 by lessw2020 Loading…
dp2ep Expert Parallel CLA Signed This label is managed by the Meta Open Source bot.
#1324 opened Jun 21, 2025 by tianyu-l Loading…
13
Support finetuning from a pretrained model CLA Signed This label is managed by the Meta Open Source bot.
#1321 opened Jun 20, 2025 by vwxyzjn Loading…
20
[not for land] testing out float8 128_1_128_128 blockwise scaling CLA Signed This label is managed by the Meta Open Source bot.
#1317 opened Jun 18, 2025 by vkuzo Loading…
Do not submit: Multinode training seems to be working CLA Signed This label is managed by the Meta Open Source bot.
#1314 opened Jun 17, 2025 by ahmadsharif1 • Draft
Do not submit: Multinode is working with multiple controllers CLA Signed This label is managed by the Meta Open Source bot.
#1313 opened Jun 17, 2025 by ahmadsharif1 • Draft
[llama4][auxiliary-loss-free load balancing] update expert_bias without backward hooks CLA Signed This label is managed by the Meta Open Source bot.
#1304 opened Jun 16, 2025 by hann-wang Loading…
Finetune from pre-trained models CLA Signed This label is managed by the Meta Open Source bot.
#1300 opened Jun 15, 2025 by vwxyzjn Loading…
25
[not for land] Use new AC CLA Signed This label is managed by the Meta Open Source bot.
#1294 opened Jun 13, 2025 by soulitzer Loading…
WIP: Try to use monarch to run torchtitan. CLA Signed This label is managed by the Meta Open Source bot.
#1288 opened Jun 12, 2025 by ahmadsharif1 • Draft
DO NOT SUBMIT: WIP: Try to use monarch to run torchtitan. CLA Signed This label is managed by the Meta Open Source bot.
#1286 opened Jun 12, 2025 by ahmadsharif1 • Draft
[deepseek][kernels][blackwell] Cutlass blackwell grouped gemm using cute dsl (forward,backward) CLA Signed This label is managed by the Meta Open Source bot.
#1276 opened Jun 8, 2025 by lessw2020 Loading…
6
[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…
ProTip! Add no:assignee to see everything that’s not assigned.