-
Notifications
You must be signed in to change notification settings - Fork 656
Pull requests: pytorch/torchtitan
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[rl] refactor model registery
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2194
opened Jan 2, 2026 by
wwwjn
Loading…
[rl] Using JobConfig as the centralized config system for inference and simple GRPO
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2191
opened Jan 2, 2026 by
wwwjn
Loading…
use comms in compiler toolkit
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
experiments: add nemotron3 model to experiments folder
CLA Signed
This label is managed by the Meta Open Source bot.
#2187
opened Dec 30, 2025 by
aghilann
Loading…
4 tasks
auto-chunk unembed & loss
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2186
opened Dec 29, 2025 by
shunting314
Loading…
[rl] Update callsite to init_batch_invariance to pass attention backend.
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2176
opened Dec 24, 2025 by
zhxchen17
Loading…
compiler_toolkit: inputs are not DTensor if TP is not enabled
CLA Signed
This label is managed by the Meta Open Source bot.
#2175
opened Dec 24, 2025 by
yanboliang
Loading…
Add Flex flash backend to flex attention module
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
[do not land] trying invoke_subgraph on torchtitan
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
[transformers_modeling_backend] Upgrade transformers from 4.57.1 to 5.0.0rc0
CLA Signed
This label is managed by the Meta Open Source bot.
#2154
opened Dec 15, 2025 by
3outeille
Loading…
[Compiler Toolkit] Add option for full inductor.
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2150
opened Dec 13, 2025 by
aditvenk
Loading…
[WIP] Use all DTensor for Qwen3 and llama4 at TP region
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2149
opened Dec 12, 2025 by
wwwjn
Loading…
Staging SFT training
CLA Signed
This label is managed by the Meta Open Source bot.
#2148
opened Dec 12, 2025 by
rakkit
Loading…
[CP] Enable FlexCP for llama3
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2145
opened Dec 11, 2025 by
fegin
Loading…
[CP] Refactor Context Parallel to use new PyTorch CP APIs
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
#2144
opened Dec 11, 2025 by
fegin
Loading…
fixed validation error when using flash attention
CLA Signed
This label is managed by the Meta Open Source bot.
#2142
opened Dec 11, 2025 by
francesco-bertolotti
Loading…
Add repeated_subgraphs option in AutoParallel example
CLA Signed
This label is managed by the Meta Open Source bot.
#2138
opened Dec 10, 2025 by
fmassa
Loading…
[Not Ready] Enable Async TP CI
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
improve throughput of HF dense model (no need actually)
CLA Signed
This label is managed by the Meta Open Source bot.
perf(pipeline): implement auto-partition algorithm
CLA Signed
This label is managed by the Meta Open Source bot.
enhancement
New feature or request
#2113
opened Dec 5, 2025 by
TXacs
Loading…
[simple_fsdp] Turn on bucketing by default
CLA Signed
This label is managed by the Meta Open Source bot.
#2103
opened Dec 3, 2025 by
IvanKobzarev
Loading…
DeepSeekV3 DualPipeV with SAC
ciflow/8gpu
CLA Signed
This label is managed by the Meta Open Source bot.
[simplefsdp] fix & enable DSV3 manual bucketing
CLA Signed
This label is managed by the Meta Open Source bot.
#2080
opened Nov 24, 2025 by
ruisizhang123
Loading…
Validate tokenizer and model alignment before training
CLA Signed
This label is managed by the Meta Open Source bot.
#2074
opened Nov 21, 2025 by
CryptoSalamander
Loading…
Previous Next
ProTip!
Type g p on any issue or pull request to go back to the pull request listing page.