Skip to content

[Performance] Qwen3-Next: speed up update_mamba_state_after_mtp_verify by 10x; e2e up to 3.54% faster #10586

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
zhyncs merged 3 commits into from
Sep 19, 2025
Merged

[Performance] Qwen3-Next: speed up update_mamba_state_after_mtp_verify by 10x; e2e up to 3.54% faster #10586

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
a78e0d4
[Performance] Qwen3-Next: speed up update_mamba_state_after_mtp_verif…
byjiang1996 Sep 18, 2025
d992257
Merge branch 'main' into byjiang1996/update_mamba_state_after_mtp_verify
zhyncs Sep 18, 2025
de0d48a
Merge branch 'main' into byjiang1996/update_mamba_state_after_mtp_verify
zhyncs Sep 19, 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
45 changes: 12 additions & 33 deletions python/sglang/srt/layers/attention/hybrid_linear_attn_backend.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -583,36 +583,15 @@ def update_mamba_state_after_mtp_verify(self, accepted_length, model):

# Compute common indices once to avoid duplication
last_steps_all = (accepted_length - 1).to(torch.int64)
valid_state_indices = state_indices_tensor[valid_mask].to(torch.int64)
last_steps = last_steps_all[valid_mask].to(torch.int64)

if valid_state_indices.numel() > 0:
chunk = 256
num_valid = valid_state_indices.numel()

# SSM state updates
for i in range(0, num_valid, chunk):
idx = valid_state_indices[i : i + chunk]
steps = last_steps[i : i + chunk]
# per (cache line, step)
for j in range(idx.numel()):
ci = idx[j].item()
st = steps[j].item()
ssm_states[:, ci, :].copy_(
intermediate_state_cache[:, ci, st].to(
ssm_states.dtype, copy=False
)
)

# Conv window updates
for i in range(0, num_valid, chunk):
idx = valid_state_indices[i : i + chunk]
steps = last_steps[i : i + chunk]
for j in range(idx.numel()):
ci = idx[j].item()
st = steps[j].item()
conv_states[:, ci, :, :].copy_(
intermediate_conv_window_cache[:, ci, st].to(
conv_states.dtype, copy=False
)
)
valid_state_indices = state_indices_tensor[valid_mask].to(torch.int64) # [N]
last_steps = last_steps_all[valid_mask].to(torch.int64) # [N]

# scatter into ssm_states at the chosen cache lines
ssm_states[:, valid_state_indices, :] = intermediate_state_cache[
:, valid_state_indices, last_steps
].to(ssm_states.dtype, copy=False)

# Scatter into conv_states at the chosen cache lines
conv_states[:, valid_state_indices, :, :] = intermediate_conv_window_cache[
:, valid_state_indices, last_steps
].to(conv_states.dtype, copy=False)
Comment thread
byjiang1996 marked this conversation as resolved.
Loading