[Spec Decode][MRV2] RecoverSSM for Qwen GDN and PLE short conv (PR #58863)
Intake of open upstream PR vllm-project/vllm#58863. Active only with VLLM_USE_V2_MODEL_RUNNER=1; inert otherwise. Speculative decoding on Gated DeltaNet snapshotted the full recurrent state after every draft position and reserved one Mamba block per draft position. On a 24 GB card that is reserved VRAM we never use, plus per-step copy traffic. RecoverSSM, the protocol Kimi-K3 KDA already uses, replaces this: verify runs from a read-only checkpoint and stores a small per-token replay record [correction (V), normalized key (K), gate (1)] in fp32; commit replays only the accepted tokens and writes the recurrent state once, plus the block-boundary state in align mode. The PLE dilated short conv compacts its accepted window with the same plan. Conflict in vllm/models/qwen4_exp/nvidia/mtp.py: our get_top_tokens() work put LocalArgmaxMixin on Qwen4ExpMTP, the PR adds SupportsReplaySSM to the same base list. Kept both. Co-Authored-By: Hinata <hinata@idihore.id> Signed-off-by: MOVZX <movzx@yahoo.com>
This commit is contained in:
@@ -261,7 +261,7 @@ def _create_decode_vllm_config(
|
||||
num_speculative_tokens: int = 0,
|
||||
dynamic_spec_schedule: list[tuple[int, int, int]] | None = None,
|
||||
max_num_seqs: int = 8,
|
||||
use_kda_recoverssm: bool = False,
|
||||
use_recoverssm: bool = False,
|
||||
) -> MagicMock:
|
||||
compilation_config = CompilationConfig(
|
||||
cudagraph_mode="FULL_AND_PIECEWISE",
|
||||
@@ -276,7 +276,7 @@ def _create_decode_vllm_config(
|
||||
max_num_seqs=max_num_seqs
|
||||
)
|
||||
vllm_config.parallel_config = ParallelConfig()
|
||||
vllm_config.cache_config = SimpleNamespace(use_kda_recoverssm=use_kda_recoverssm)
|
||||
vllm_config.cache_config = SimpleNamespace(use_recoverssm=use_recoverssm)
|
||||
vllm_config.num_speculative_tokens = num_speculative_tokens
|
||||
if dynamic_spec_schedule is None:
|
||||
vllm_config.speculative_config = None
|
||||
@@ -295,7 +295,7 @@ def _make_spec_decode_manager(
|
||||
num_speculative_tokens: int = 0,
|
||||
dynamic_spec_schedule: list[tuple[int, int, int]] | None = None,
|
||||
max_num_seqs: int = 8,
|
||||
use_kda_recoverssm: bool = False,
|
||||
use_recoverssm: bool = False,
|
||||
varlen_decode: bool = False,
|
||||
) -> gpu_cudagraph_utils.CudaGraphManager:
|
||||
monkeypatch.setattr(
|
||||
@@ -314,7 +314,7 @@ def _make_spec_decode_manager(
|
||||
num_speculative_tokens=num_speculative_tokens,
|
||||
dynamic_spec_schedule=dynamic_spec_schedule,
|
||||
max_num_seqs=max_num_seqs,
|
||||
use_kda_recoverssm=use_kda_recoverssm,
|
||||
use_recoverssm=use_recoverssm,
|
||||
),
|
||||
device=torch.device("cpu"),
|
||||
cudagraph_mode=CUDAGraphMode.FULL_AND_PIECEWISE,
|
||||
|
||||
Reference in New Issue
Block a user