[CI/Build] Split up models tests (#10069)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

tests/models/decoder_only/language/test_models.py

@@ -1,8 +1,5 @@
 """Compare the outputs of HF and vLLM when using greedy sampling.
 
-This test only tests small models. Big models such as 7B should be tested from
-test_big_models.py because it could use a larger instance to run tests.
-
 Run `pytest tests/models/test_models.py`.
 """
 import pytest
@@ -35,6 +32,7 @@ if not current_platform.is_cpu():
 target_dtype = "half"
 
 
+@pytest.mark.core_model
 @pytest.mark.parametrize("model", MODELS)
 @pytest.mark.parametrize("dtype", [target_dtype])
 @pytest.mark.parametrize("max_tokens", [32])