diff --git a/src/wikiq/__init__.py b/src/wikiq/__init__.py
index 9dcc64b..37f2e22 100755
--- a/src/wikiq/__init__.py
+++ b/src/wikiq/__init__.py
@@ -3,8 +3,8 @@
 # original wikiq headers are: title articleid revid date_time anon
 # editor editor_id minor text_size text_entropy text_md5 reversion
 # additions_size deletions_size
-import gc
 import argparse
+import gc
 import json
 import os.path
 import re
@@ -465,7 +465,6 @@ class WikiqParser:
             next_batch = {}
             diff_dict = {}
 
-
             if self.persist != PersistMethod.none:
                 window = deque(maxlen=PERSISTENCE_RADIUS)
                 if self.persist != PersistMethod.none:
@@ -488,11 +487,14 @@ class WikiqParser:
                         from mw.lib import persistence
 
                         persist_state = persistence.State()
-
-
+q
             if self.diff:
                 differ = pywikidiff2.pywikidiff2(
-                    numContextLines=1000000, moved_paragraph_detection_cutoff=2000
+                    numContextLines=1000000,
+                    moved_paragraph_detection_cutoff=200000,
+                    words_cache_capacity=10000,
+                    diff_cache_capacity=10000,
+                    stats_cache_capacity=100000,
                 )
 
             while not on_last_batch:
@@ -595,7 +597,6 @@ class WikiqParser:
                     row_buffer[k] = v
                     regex_matches = {}
 
-
                 # begin persistence logic
                 if self.persist != PersistMethod.none:
                     row_buffer["token_revs"] = []
@@ -937,7 +938,7 @@ def main():
                 diff=args.diff,
                 output_parquet=output_parquet,
                 partition_namespaces=args.partition_namespaces,
-                batch_size = args.batch_size
+                batch_size=args.batch_size,
             )
 
             wikiq.process()
@@ -960,7 +961,7 @@ def main():
             regex_comment_label=args.regex_comment_label,
             diff=args.diff,
             text=args.text,
-            batch_size=args.batch_size
+            batch_size=args.batch_size,
         )
 
         wikiq.process()
diff --git a/src/wikiq/wiki_diff_matcher.py b/src/wikiq/wiki_diff_matcher.py
index 6dfdab6..f9b8234 100644
--- a/src/wikiq/wiki_diff_matcher.py
+++ b/src/wikiq/wiki_diff_matcher.py
@@ -330,15 +330,17 @@ class WikiDiffMatcher:
     ):
         self.tokenizer = tokenizer or TOKENIZER
 
-            
-
     class Processor(DiffEngine.Processor):
         def __init__(self, tokenizer=None):
             self.tokenizer = tokenizer or TOKENIZER
             self.last_tokens = []
             self.previous_text = ""
             self.differ = pywikidiff2.pywikidiff2(
-                numContextLines=1000000, moved_paragraph_detection_cutoff=200000
+                numContextLines=1000000,
+                moved_paragraph_detection_cutoff=200000,
+                words_cache_capacity=10000,
+                diff_cache_capacity=10000,
+                stats_cache_capacity=100000,
             )
             self.last_diff = None