diff --git a/api/routes.py b/api/routes.py
index 3e5ae1bc..eff7e3c3 100644
--- a/api/routes.py
+++ b/api/routes.py
@@ -101,6 +101,8 @@ def _sync_session_title_to_insights(session) -> None:
             title=session.title,
             message_count=len(messages),
             profile=getattr(session, "profile", None),
+            cache_read_tokens=getattr(session, "cache_read_tokens", None) or 0,
+            cache_write_tokens=getattr(session, "cache_write_tokens", None) or 0,
         )
     except Exception:
         logger.debug("Failed to update session title in state.db", exc_info=True)
@@ -20690,6 +20692,8 @@ def _handle_chat_sync(handler, body):
                 model=s.model,
                 title=s.title,
                 message_count=len(s.messages),
+                cache_read_tokens=s.cache_read_tokens or 0,
+                cache_write_tokens=s.cache_write_tokens or 0,
                 # #2762 / #2827 parity with api/streaming.py:5078: pass the
                 # session's profile explicitly so a future refactor that
                 # backgrounds this handler doesn't silently leak writes to
diff --git a/api/state_sync.py b/api/state_sync.py
index b35a57bf..07cc1ccb 100644
--- a/api/state_sync.py
+++ b/api/state_sync.py
@@ -134,7 +134,8 @@ def sync_session_start(session_id: str, model=None, profile: Optional[str] = Non
 
 def sync_session_usage(session_id: str, input_tokens: int=0, output_tokens: int=0,
                        estimated_cost=None, model=None, title: Optional[str] = None,
-                       message_count: Optional[int] = None, profile: Optional[str] = None) -> None:
+                       message_count: Optional[int] = None, profile: Optional[str] = None,
+                       cache_read_tokens: int = 0, cache_write_tokens: int = 0) -> None:
     """Update token usage and title for a WebUI session in state.db.
     Called after each turn completes. Uses absolute=True to set totals
     (the WebUI Session already accumulates across turns).
@@ -153,11 +154,17 @@ def sync_session_usage(session_id: str, input_tokens: int=0, output_tokens: int=
     try:
         # Ensure session exists first (idempotent)
         db.ensure_session(session_id=session_id, source='webui', model=model)
-        # Set absolute token counts
+        # Set absolute token counts. WebUI's sidecar already accumulates
+        # input/output/cache totals across turns, so mirror the same absolute
+        # values into state.db. Omitting cache counters makes insights/reporting
+        # show false 0% hit rates even when the live stream and sidecar saw warm
+        # prefix reads.
         db.update_token_counts(
             session_id=session_id,
             input_tokens=input_tokens,
             output_tokens=output_tokens,
+            cache_read_tokens=cache_read_tokens,
+            cache_write_tokens=cache_write_tokens,
             estimated_cost_usd=estimated_cost,
             model=model,
             absolute=True,
diff --git a/api/streaming.py b/api/streaming.py
index 0fd7a829..6b32c7f4 100644
--- a/api/streaming.py
+++ b/api/streaming.py
@@ -9287,6 +9287,8 @@ def _run_agent_streaming(
                             model=model,
                             title=s.title,
                             message_count=len(s.messages),
+                            cache_read_tokens=s.cache_read_tokens or 0,
+                            cache_write_tokens=s.cache_write_tokens or 0,
                             # #2762: pass the session's profile explicitly so the
                             # background-thread state.db lookup doesn't fall
                             # through to the process-global active profile and
diff --git a/tests/test_issue2762_state_sync_profile_kwarg.py b/tests/test_issue2762_state_sync_profile_kwarg.py
index b6f82f30..a420e9dc 100644
--- a/tests/test_issue2762_state_sync_profile_kwarg.py
+++ b/tests/test_issue2762_state_sync_profile_kwarg.py
@@ -82,12 +82,14 @@ def _read_session(db_path: Path, session_id: str):
     if not db_path.exists():
         return None
     conn = sqlite3.connect(db_path)
+    conn.row_factory = sqlite3.Row
     try:
         # Real state.db schema (see api/state_sync.py + hermes_cli StateDB):
         # `sessions` table has `id` as PRIMARY KEY (not session_id). Use real
         # column names so the test queries the actual schema.
         cur = conn.execute(
-            "SELECT id AS session_id, title, input_tokens, output_tokens "
+            "SELECT id AS session_id, title, input_tokens, output_tokens, "
+            "cache_read_tokens, cache_write_tokens "
             "FROM sessions WHERE id = ?",
             (session_id,),
         )
@@ -213,6 +215,8 @@ def test_sync_session_usage_writes_only_to_named_profile(two_profile_homes):
         title='2762 regression test',
         message_count=3,
         profile='maiko',
+        cache_read_tokens=1200,
+        cache_write_tokens=300,
     )
 
     maiko_row = _read_session(two_profile_homes['maiko'] / 'state.db', '2762-regression')
@@ -220,6 +224,8 @@ def test_sync_session_usage_writes_only_to_named_profile(two_profile_homes):
 
     assert maiko_row is not None, \
         "sync_session_usage(profile='maiko') did not write to maiko's state.db"
+    assert maiko_row["cache_read_tokens"] == 1200
+    assert maiko_row["cache_write_tokens"] == 300
     assert hiyuki_row is None, \
         "sync_session_usage(profile='maiko') leaked into hiyuki's state.db — #2762 regression"