perf-sequence-packing

Original：🇺🇸 English
Translated

Validate and use packed sequences and long-context training in Megatron-Bridge, distinguishing offline packed SFT for LLMs from in-batch packing for VLMs, and applying the right CP constraints.

4installs
Sourcenvidia/skills
Added on

NPX Install

npx skill4agent add nvidia/skills perf-sequence-packing

Tags

Translated version includes tags in frontmatter
megatron-bridgesequence-packinglong-context-trainingllm-finetuningvlm-finetuning

SKILL.md Content

View Translation Comparison →

Sequence Packing Skill

For stable background and recommendation level, see:
  • @docs/training/packed-sequences.md
  • @skills/perf-sequence-packing/card.yaml

Enablement

Offline packed SFT for LLM finetuning:
python
from megatron.bridge.data.datasets.packed_sequence import PackedSequenceSpecs

cfg.train.micro_batch_size = 1
cfg.dataset.seq_length = 4096
cfg.model.seq_length = 4096
cfg.dataset.dataset_kwargs = {"pad_to_max_length": True}
cfg.dataset.packed_sequence_specs = PackedSequenceSpecs(
    packed_sequence_size=4096,
    pad_seq_to_mult=1,
)
If CP is enabled:
python
cfg.model.context_parallel_size = 2
cfg.model.calculate_per_token_loss = True
cfg.ddp.average_in_collective = False
cfg.dataset.packed_sequence_specs.pad_seq_to_mult = cfg.model.context_parallel_size * 2

# If sequence_parallel is also enabled, use lcm(2*CP, CP*TP):
# import math
# cfg.dataset.packed_sequence_specs.pad_seq_to_mult = math.lcm(2 * CP, CP * TP)
# See src/megatron/bridge/training/vlm_step.py for reference logic.
If CUDA graphs are enabled for this packed path:
python
cfg.dataset.packed_sequence_specs.pad_cu_seqlens = True
cfg.dataset.dataset_kwargs["pad_to_max_length"] = True
Note: 
pad_cu_seqlens = True
also requires a metadata JSON file alongside the packed dataset (asserted in 
src/megatron/bridge/data/datasets/sft.py
). Custom packed datasets that omit the metadata file will hit an assertion at dataset initialization.
In-batch packing for VLM finetuning:
python
cfg.dataset.pack_sequences_in_batch = True
cfg.train.micro_batch_size = 2
Long-context baseline:
python
cfg.model.seq_length = 16384
cfg.dataset.seq_length = 16384
cfg.model.context_parallel_size = 2

Code Anchors

LLM packed SFT config surface:
72
if packed_sequence:
    dataset_kwargs = {"pad_to_max_length": True}
    packed_sequence_specs = PackedSequenceSpecs(packed_sequence_size=seq_length, pad_seq_to_mult=pad_seq_to_mult)
else:
    dataset_kwargs = {}
    packed_sequence_specs = None
Bridge validation:
1617
if self.model.context_parallel_size > 1:
    assert self.model.seq_length % (self.model.context_parallel_size * 2) == 0, ...
    if isinstance(self.dataset, FinetuningDatasetConfig):
        assert self.model.calculate_per_token_loss, ...
        assert not self.ddp.average_in_collective, ...
...
if ... packed_sequence_size > 0 and self.train.micro_batch_size > 1:
    raise ValueError(...)
...
if getattr(self.dataset, "pack_sequences_in_batch", False) and self.train.micro_batch_size == 1:
    raise ValueError(...)
VLM in-batch runtime:
308
if enable_packing:
    ...
    ) = pack_batch_sequences(
        ...
        pad_token_id=0,
        pad_to_multiple_of=cp_size * 2 if cp_size > 1 else 1,
    )
Packed THD runtime constraint:
61
if cu_seqlens.dim() > 1 and cu_seqlens.size(0) != 1:
    raise ValueError("Packed THD batches expect micro-batch size 1 for context-parallel slicing (THD layout)")

Pitfalls

  1. Offline packed SFT and VLM in-batch packing are different features with opposite micro-batch rules.
  2. When CP is enabled, packed sequence lengths must respect 
    2 * context_parallel_size
    divisibility.
  3. For finetuning with CP, 
    calculate_per_token_loss=True
    and 
    ddp.average_in_collective=False
    are required.
  4. pad_cu_seqlens=True
    also requires 
    pad_to_max_length=True
    .
  5. Packing support is model-family-specific. 
    Qwen3-Next
    , 
    GLM-4.5
    , and 
    Qwen3.5-VL
    contain explicit opt-outs in different paths.
  6. MTP finetuning is documented as incompatible with packed sequences.

Verification

Use the checked-in unit coverage:
bash
uv run python -m pytest tests/unit_tests/training/utils/test_packed_seq_utils.py -v && \
uv run python -m pytest tests/unit_tests/training/test_config.py -k "packed_sequence or pack_sequences_in_batch or context_parallel_seq_length_divisibility or context_parallel_finetuning_validations" -v && \
uv run python -m pytest tests/unit_tests/training/test_vlm_step.py -k "enable_packing" -v
Success criteria:
  • first command reports 
    8 passed
  • second command reports 
    14 passed
  • third command reports 
    2 passed