Skip to content

[model] refactor: polishing FSDP model engine #3394

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 9 commits into from
Sep 8, 2025
Merged

[model] refactor: polishing FSDP model engine #3394

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
9 commits
Select commit Hold shift + click to select a range
45cf91d
polish fsdp engine
vermouth1992 Sep 8, 2025
52c71fc
update
vermouth1992 Sep 8, 2025
b71ceca
precommit
vermouth1992 Sep 8, 2025
c71fb8d
precommit
vermouth1992 Sep 8, 2025
3fcdd41
toggle out fsdp2 test
vermouth1992 Sep 8, 2025
112a3e2
precommit
vermouth1992 Sep 8, 2025
2db98f9
update
vermouth1992 Sep 8, 2025
d63fdde
fix sp size
vermouth1992 Sep 8, 2025
eabff29
fix vlm
vermouth1992 Sep 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 6 additions & 2 deletions tests/special_e2e/sft/test_sft_engine_all.sh
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,12 +18,16 @@ BACKEND=fsdp SP_SIZE=2 FSDP_SIZE=-1 NUM_GPUS=8 FSDP_STRATEGY=fsdp bash tests/spe
echo "run with sp4 fsdp_size4 num_gpus8 fsdp_strategy fsdp"
BACKEND=fsdp SP_SIZE=4 FSDP_SIZE=4 NUM_GPUS=8 FSDP_STRATEGY=fsdp bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh

# test with fsdp 2
echo "run with sp1 fsdp_size1 num_gpus1 fsdp_strategy fsdp2"
BACKEND=fsdp SP_SIZE=1 FSDP_SIZE=1 NUM_GPUS=1 FSDP_STRATEGY=fsdp2 bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh

# TODO: toggle the follow tests when the grad norm of fsdp is fixed
# BACKEND=fsdp SP_SIZE=1 FSDP_SIZE=1 NUM_GPUS=1 FSDP_STRATEGY=fsdp2 bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh
# BACKEND=fsdp SP_SIZE=1 FSDP_SIZE=2 NUM_GPUS=8 FSDP_STRATEGY=fsdp2 bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh
# echo "run with sp1 fsdp_size-1 num_gpus8 fsdp_strategy fsdp2"
# BACKEND=fsdp SP_SIZE=1 FSDP_SIZE=-1 NUM_GPUS=8 FSDP_STRATEGY=fsdp2 bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh
# echo "run with sp2 fsdp_size-1 num_gpus8 fsdp_strategy fsdp2"
# BACKEND=fsdp SP_SIZE=2 FSDP_SIZE=-1 NUM_GPUS=8 FSDP_STRATEGY=fsdp2 bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh
# BACKEND=fsdp SP_SIZE=1 FSDP_SIZE=2 NUM_GPUS=8 FSDP_STRATEGY=fsdp2 bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh
# BACKEND=fsdp SP_SIZE=4 FSDP_SIZE=4 NUM_GPUS=8 FSDP_STRATEGY=fsdp2 bash tests/special_e2e/sft/run_sft_engine_gsm8k.sh

# test with megatron
Expand Down
3 changes: 2 additions & 1 deletion verl/utils/seqlen_balancing.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -354,7 +354,8 @@ def prepare_dynamic_batch(data: DataProto, max_token_len: int) -> tuple[list[Dat
for i, batch_idx in enumerate(batch_idx_list):
tensors = dict(batch[i])
non_tensors = {key: value[batch_idx] for key, value in data.non_tensor_batch.items()}
micro_batches.append(DataProto.from_dict(tensors, non_tensors, meta_info=data.meta_info))
meta_info = copy.deepcopy(data.meta_info)
micro_batches.append(DataProto.from_dict(tensors, non_tensors, meta_info=meta_info))

return micro_batches, batch_idx_list

Expand Down
Loading
Loading