Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
change
sglang
Commits
42e1a72e
Unverified
Commit
42e1a72e
authored
Oct 28, 2025
by
hlu1
Committed by
GitHub
Oct 28, 2025
Browse files
[Deepseek V3.2] Enable flashmla_auto with MTP (#12294)
Signed-off-by:
Hao Lu
<
14827759+hlu1@users.noreply.github.com
>
parent
83a7c89c
Changes
1
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
1 addition
and
3 deletions
+1
-3
python/sglang/srt/layers/attention/nsa_backend.py
python/sglang/srt/layers/attention/nsa_backend.py
+1
-3
No files found.
python/sglang/srt/layers/attention/nsa_backend.py
View file @
42e1a72e
...
...
@@ -1222,11 +1222,9 @@ class NativeSparseAttnBackend(AttentionBackend):
if
self
.
enable_auto_select_prefill_impl
:
if
self
.
nsa_kv_cache_store_fp8
:
if
(
# TODO(hlu1): enable MTP
is_blackwell
()
and
forward_batch
is
not
None
and
forward_batch
.
forward_mode
.
is_extend
()
and
forward_batch
.
spec_algorithm
.
is_none
()
and
forward_batch
.
forward_mode
==
ForwardMode
.
EXTEND
):
total_kv_tokens
=
forward_batch
.
seq_lens_sum
total_q_tokens
=
forward_batch
.
extend_num_tokens
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
.
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment