[CI Failure] Fix tests with missing TinyLlama-1.1B-Chat-v1.0-FP8-e2e (#26816)

Signed-off-by: mgoin <mgoin64@gmail.com>

tests/compile/test_async_tp.py

@@ -332,7 +332,7 @@ def async_tp_pass_on_test_model(
 
     # this is a fake model name to construct the model config
     # in the vllm_config, it's not really used.
-    model_name = "nm-testing/TinyLlama-1.1B-Chat-v1.0-FP8-e2e"
+    model_name = "RedHatAI/Llama-3.2-1B-Instruct-FP8"
     vllm_config.model_config = ModelConfig(
         model=model_name, trust_remote_code=True, dtype=dtype, seed=42
     )

tests/compile/test_fusion_all_reduce.py

@@ -229,7 +229,7 @@ def all_reduce_fusion_pass_on_test_model(
 
     # this is a fake model name to construct the model config
     # in the vllm_config, it's not really used.
-    model_name = "nm-testing/TinyLlama-1.1B-Chat-v1.0-FP8-e2e"
+    model_name = "RedHatAI/Llama-3.2-1B-Instruct-FP8"
     vllm_config.model_config = ModelConfig(
         model=model_name, trust_remote_code=True, dtype=dtype, seed=42
     )

tests/compile/test_sequence_parallelism.py

@@ -278,7 +278,7 @@ def sequence_parallelism_pass_on_test_model(
 
     # this is a fake model name to construct the model config
     # in the vllm_config, it's not really used.
-    model_name = "nm-testing/TinyLlama-1.1B-Chat-v1.0-FP8-e2e"
+    model_name = "RedHatAI/Llama-3.2-1B-Instruct-FP8"
     vllm_config.model_config = ModelConfig(
         model=model_name, trust_remote_code=True, dtype=dtype, seed=42
     )