From 56c93ab89ea10c63abba9f9e5154050b742fc5da Mon Sep 17 00:00:00 2001
From: "google-labs-jules[bot]"
 <161369871+google-labs-jules[bot]@users.noreply.github.com>
Date: Tue, 5 May 2026 15:57:17 +0000
Subject: [PATCH] =?UTF-8?q?=E2=9A=A1=20Bolt:=20[performance=20improvement]?=
 =?UTF-8?q?=20Replace=20O(N^2)=20list=20pop(0)=20in=20entropy=20calculatio?=
 =?UTF-8?q?n?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .jules/bolt.md                             |  3 +++
 fastdeploy/model_executor/entropy_utils.py | 16 ++++++++++++----
 2 files changed, 15 insertions(+), 4 deletions(-)
 create mode 100644 .jules/bolt.md

diff --git a/.jules/bolt.md b/.jules/bolt.md
new file mode 100644
index 00000000000..e85ac3ab0d3
--- /dev/null
+++ b/.jules/bolt.md
@@ -0,0 +1,3 @@
+## 2025-05-05 - Replacing O(N^2) list pop(0) in fastdeploy/model_executor/entropy_utils.py
+**Learning:** Found O(N^2) list operations where `entropy.pop(0)` was called inside nested loops over batch size and sequence lengths.
+**Action:** Replace `pop(0)` within a loop with an index tracker and bulk addition via `.extend(list[start:end])`.
diff --git a/fastdeploy/model_executor/entropy_utils.py b/fastdeploy/model_executor/entropy_utils.py
index 21d1b3421e9..9fd2605775f 100644
--- a/fastdeploy/model_executor/entropy_utils.py
+++ b/fastdeploy/model_executor/entropy_utils.py
@@ -49,9 +49,13 @@ def calculate_logits_entropy(logits, share_inputs, temperature):
     entropy_tensor = get_entropy(logits)
     entropy = entropy_tensor.tolist()
 
+    entropy_idx = 0
     for i in range(real_bsz):
-        for _ in range(real_seq_lens[i]):
-            share_inputs["entropy_list"][i].append(entropy.pop(0))
+        seq_len = int(real_seq_lens[i])
+        if seq_len > 0:
+            share_inputs["entropy_list"][i].extend(entropy[entropy_idx : entropy_idx + seq_len])
+            entropy_idx += seq_len
+
         if (
             share_inputs["stop_flags"][i]
             and share_inputs["seq_lens_decoder"][i] != 0
@@ -92,9 +96,13 @@ def speculate_calculate_logits_entropy(logits, share_inputs, temperature):
     entropy_tensor = get_entropy(accepted_logits)
     entropy = entropy_tensor.tolist()
 
+    entropy_idx = 0
     for i in range(real_bsz):
-        for _ in range(share_inputs["accept_num"][i]):
-            share_inputs["entropy_list"][i].append(entropy.pop(0))
+        accept_num = int(share_inputs["accept_num"][i])
+        if accept_num > 0:
+            share_inputs["entropy_list"][i].extend(entropy[entropy_idx : entropy_idx + accept_num])
+            entropy_idx += accept_num
+
         if (
             share_inputs["stop_flags"][i]
             and share_inputs["seq_lens_decoder"][i] != 0