[Perf][MRV2] Allow FULL decode graphs for one-token prompt tails (#58400)

Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Nicolò Lucchesi <nicolo.lucchesi@mistral.ai>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Yongye Zhu <zyy1102000@gmail.com>
This commit is contained in:
Nick Hill
2026-09-28 12:48:52 -07:00
committed by GitHub
co-authored by Nicolò Lucchesi Claude Opus 5.5 mergify[bot] <37929162+mergify[bot]@users.noreply.github.com> Yongye Zhu
parent ef42093a36
commit e7c903609f
12 changed files with 144 additions and 42 deletions
+4 -2
View File
@@ -747,10 +747,12 @@ def is_uniform_query_len(num_reqs: int, num_tokens: int, max_query_len: int) ->
def get_uniform_decode_token_count(
num_reqs: int, num_tokens: int, max_query_len: int, has_prefill: bool
num_reqs: int, num_tokens: int, max_query_len: int, decode_graph_eligible: bool
) -> int | None:
"""Per-request token count of a uniform decode batch, or None."""
if not has_prefill and is_uniform_query_len(num_reqs, num_tokens, max_query_len):
if decode_graph_eligible and is_uniform_query_len(
num_reqs, num_tokens, max_query_len
):
return max_query_len
return None