Fix ExaoneMoeMTP test that never ran in Transformers v4 (#36792)

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

tests/distributed/test_pipeline_parallel.py

@@ -247,6 +247,7 @@ def _compare_tp(
     hf_config = get_config(model_id, trust_remote_code)
     require_embed_inputs = model_info.require_embed_inputs
     max_num_seqs = model_info.max_num_seqs
+    enable_prefix_caching = model_info.enable_prefix_caching
 
     dtype = "float16"
     if hf_config.model_type in _FLOAT16_NOT_SUPPORTED_MODELS:
@@ -300,6 +301,8 @@ def _compare_tp(
         common_args.extend(["--load-format", load_format])
     if hf_overrides:
         common_args.extend(["--hf-overrides", json.dumps(hf_overrides)])
+    if not enable_prefix_caching:
+        common_args.append("--no-enable-prefix-caching")
     if require_embed_inputs:
         common_args.extend(
             [

tests/models/multimodal/generation/vlm_utils/core.py

@@ -74,6 +74,8 @@ def run_test(
     if model_info.require_embed_inputs:
         for k in ("skip_tokenizer_init", "enable_prompt_embeds", "enable_mm_embeds"):
             vllm_runner_kwargs_[k] = model_info.require_embed_inputs
+    if not model_info.enable_prefix_caching:
+        vllm_runner_kwargs_["enable_prefix_caching"] = False
 
     if vllm_runner_kwargs:
         vllm_runner_kwargs_.update(vllm_runner_kwargs)

tests/models/registry.py

@@ -72,6 +72,12 @@ class _HfExamplesInfo:
     If False, we will use CUDA graph and eager execution in hybrid.
     """
 
+    enable_prefix_caching: bool = True
+    """
+    Whether to enable prefix caching for the model. If True, we will test the model with
+    prefix caching enabled. If False, we will test the model without prefix caching.
+    """
+
     is_available_online: bool = True
     """
     Set this to `False` if the name of this architecture no longer exists on
@@ -1206,6 +1212,7 @@ _SPECULATIVE_DECODING_EXAMPLE_MODELS = {
         "LGAI-EXAONE/K-EXAONE-236B-A23B",
         speculative_model="LGAI-EXAONE/K-EXAONE-236B-A23B",
         min_transformers_version="5.1.0",
+        enable_prefix_caching=False,
     ),
     "ExtractHiddenStatesModel": _HfExamplesInfo(
         "Qwen/Qwen3-8B",

tests/models/test_initialization.py

@@ -136,6 +136,10 @@ def can_initialize(
         if model_arch == "WhisperForConditionalGeneration":
             m.setenv("VLLM_WORKER_MULTIPROC_METHOD", "spawn")
 
+        kwargs = {}
+        if not model_info.enable_prefix_caching:
+            kwargs["enable_prefix_caching"] = False
+
         LLM(
             model_info.default,
             tokenizer=model_info.tokenizer,
@@ -165,6 +169,7 @@ def can_initialize(
             hf_overrides=hf_overrides_fn,
             max_num_seqs=model_info.max_num_seqs,
             attention_config=attention_config,
+            **kwargs,
         )