diff --git a/litellm/proxy/client/cli/README.md b/litellm/proxy/client/cli/README.md
index cb867cf9e61..a5a2675ed6e 100644
--- a/litellm/proxy/client/cli/README.md
+++ b/litellm/proxy/client/cli/README.md
@@ -580,8 +580,8 @@ What the command changed is recorded in `~/.litellm/claude_configure_state.json`
 `lite configure claude`, `lite login --config-claude`, `lite up` and `lite autoroute up` also install a status line (`~/.litellm/statusline.py`, registered as `statusLine` in `~/.claude/settings.json` unless you already run one) that shows which model the auto-router actually served the last turn and, once the proxy has recorded the session, what the session cost against the router's savings baseline:
 
 ```
-claude-auto · Routed to: claude-haiku-4-5  -63% vs Claude Opus 5
-LiteLLM       ████████░░░░░░░░░░░░░░░░  $0.14
+Routed to: claude-haiku-4-5  -63% vs Claude Opus 5
+claude-auto   ████████░░░░░░░░░░░░░░░░  $0.14
 Claude Opus 5 ████████████████████████  $0.38
 ```
 
diff --git a/litellm/proxy/client/cli/commands/statusline_script.py b/litellm/proxy/client/cli/commands/statusline_script.py
index 815875d59b7..47be3888a58 100644
--- a/litellm/proxy/client/cli/commands/statusline_script.py
+++ b/litellm/proxy/client/cli/commands/statusline_script.py
@@ -43,7 +43,6 @@ FETCH_TIMEOUT_SECONDS: Final = 3
 BAR_WIDTH: Final = 24
 BAR_FULL: Final = "\u2588"
 BAR_EMPTY: Final = "\u2591"
-SEPARATOR: Final = " \u00b7 "
 TRANSCRIPT_SCAN_LIMIT_BYTES: Final = 4 * 1024 * 1024
 CLAUDE_BASE_URL_ENV_KEYS: Final = ("ANTHROPIC_BASE_URL",)
 CLAUDE_API_KEY_ENV_KEYS: Final = ("ANTHROPIC_AUTH_TOKEN", "ANTHROPIC_API_KEY")
@@ -315,11 +314,8 @@ def render(model: str, session: Session | None, config_dir: Path, use_color: boo
         return f"{code}{text}{RESET}" if use_color else text
 
     routed: Final = paint(BOLD, f"Routed to: {model}")
-    if session is None:
+    if session is None or session.baseline_model is None or session.baseline_spend <= 0:
         return routed
-    header: Final = f"{session.router_name}{SEPARATOR}{routed}"
-    if session.baseline_model is None or session.baseline_spend <= 0:
-        return header
     reference: Final = baseline_label(session.baseline_model, config_dir)
     pct: Final = (session.baseline_spend - session.spend) / session.baseline_spend * 100
     delta: Final = paint(LITELLM_COLOR, f"{'-' if pct >= 0 else '+'}{abs(round(pct))}% vs {reference}")
@@ -335,7 +331,7 @@ def render(model: str, session: Session | None, config_dir: Path, use_color: boo
         f"{paint(DIM, f'${amount:.2f}')}"
         for label, amount, color in rows
     )
-    return "\n".join((f"{header}  {delta}", *lines))
+    return "\n".join((f"{routed}  {delta}", *lines))
 
 
 def color_enabled(env: Mapping[str, str]) -> bool:
diff --git a/tests/test_litellm/proxy/client/cli/test_statusline_script.py b/tests/test_litellm/proxy/client/cli/test_statusline_script.py
index 122c9601c33..39d0e24d7b0 100644
--- a/tests/test_litellm/proxy/client/cli/test_statusline_script.py
+++ b/tests/test_litellm/proxy/client/cli/test_statusline_script.py
@@ -252,7 +252,7 @@ class TestRender:
     def test_savings_header_and_bars_against_the_routers_baseline(self, config_dir):
         text = render("claude-sonnet-5", RECORDED, config_dir, use_color=False, bar_width=10)
         assert text.splitlines() == [
-            "claude-auto · Routed to: claude-sonnet-5  -63% vs Claude Opus 5",
+            "Routed to: claude-sonnet-5  -63% vs Claude Opus 5",
             "claude-auto   ████░░░░░░ $0.14",
             "Claude Opus 5 ██████████ $0.38",
         ]
@@ -330,7 +330,7 @@ class TestRender:
         assert "+25% vs Claude Opus 5" in render("m", dearer, config_dir, use_color=False)
 
     def test_without_a_baseline_only_the_routed_line_shows(self, config_dir):
-        assert render("m", RECORDED._replace(baseline_model=None), config_dir, False) == "claude-auto · Routed to: m"
+        assert render("m", RECORDED._replace(baseline_model=None), config_dir, False) == "Routed to: m"
         assert render("m", None, config_dir, False) == "Routed to: m"
 
     def test_color_wraps_the_same_text(self, config_dir):
@@ -352,7 +352,7 @@ class TestClaudeCodeMode:
             return Fetched(RECORDED, definitive=True)
 
         text: Final = _run(_payload(transcript), _env(tmp_path, config_dir), fetch)
-        assert text.startswith("claude-auto · Routed to: claude-sonnet-5  -63% vs Claude Opus 5\n")
+        assert text.startswith("Routed to: claude-sonnet-5  -63% vs Claude Opus 5\n")
         assert text.splitlines()[1].startswith("claude-auto ")
 
     def test_a_discovered_display_name_labels_the_sessions_model(
@@ -364,7 +364,7 @@ class TestClaudeCodeMode:
             return Fetched(session, definitive=True)
 
         text: Final = _run(_payload(transcript), _env(tmp_path, config_dir), fetch)
-        assert text.startswith("claude-auto · Routed to: Claude Opus 5  -63% vs Claude Opus 5\n")
+        assert text.startswith("Routed to: Claude Opus 5  -63% vs Claude Opus 5\n")
 
     def test_an_unrecorded_session_degrades_to_the_routed_line(self, tmp_path, transcript, config_dir):
         assert _run(_payload(transcript), _env(tmp_path, config_dir), lambda c, s: Fetched(None, True)) == (
@@ -420,7 +420,7 @@ class TestCodexMode:
 
         out = _run({"hook_event_name": "Stop", "session_id": SESSION_ID, "transcript_path": "/nope"}, env, fetch)
         message = json.loads(out)["systemMessage"]
-        assert message.splitlines()[1] == "claude-auto · Routed to: claude-sonnet-5  -63% vs Claude Opus 5"
+        assert message.splitlines()[1] == "Routed to: claude-sonnet-5  -63% vs Claude Opus 5"
         assert message.splitlines()[2].startswith("claude-auto ")
         assert message.startswith("\n")
         assert seen == [Credentials("http://127.0.0.1:4000", "sk-codex")]