diff --git a/fire_planner/examples/llm_extract.py b/fire_planner/examples/llm_extract.py
index c6e7ca1..22c51cf 100644
--- a/fire_planner/examples/llm_extract.py
+++ b/fire_planner/examples/llm_extract.py
@@ -27,7 +27,7 @@ log = logging.getLogger(__name__)
 # in the constant is historical — the value can be any llama-swap model id
 # (e.g. `qwen3vl-4b` when k8s-node1's VRAM is mostly held by immich-ml).
 QWEN_MODEL = os.environ.get("LLM_MODEL", "qwen3-8b")
-CLAUDE_AGENT_MODEL = "claude-haiku-4-5"
+CLAUDE_AGENT_MODEL = "haiku"
 HTTP_TIMEOUT = httpx.Timeout(60.0)
 
 PROMPT_SYSTEM = (
diff --git a/tests/test_examples_llm_extract.py b/tests/test_examples_llm_extract.py
index a28c4f9..ab9026c 100644
--- a/tests/test_examples_llm_extract.py
+++ b/tests/test_examples_llm_extract.py
@@ -110,7 +110,7 @@ async def test_fallback_escalates_when_qwen_returns_none() -> None:
         )
 
     assert out is not None
-    assert out.llm_model == "claude-haiku-4-5"
+    assert out.llm_model == "haiku"
     assert out.country == "Philippines"
 
 
@@ -140,7 +140,7 @@ async def test_fallback_escalates_on_low_confidence() -> None:
 
     assert out is not None
     assert out.country == "Thailand"
-    assert out.llm_model == "claude-haiku-4-5"
+    assert out.llm_model == "haiku"
 
 
 @respx.mock