[Deprecation][2/N] Replace --task with --runner and --convert (#21470)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

examples/offline_inference/basic/classify.py

@@ -12,7 +12,9 @@ def parse_args():
     parser = EngineArgs.add_cli_args(parser)
     # Set example specific arguments
     parser.set_defaults(
-        model="jason9693/Qwen2.5-1.5B-apeach", task="classify", enforce_eager=True
+        model="jason9693/Qwen2.5-1.5B-apeach",
+        runner="pooling",
+        enforce_eager=True,
     )
     return parser.parse_args()
 
@@ -27,7 +29,7 @@ def main(args: Namespace):
     ]
 
     # Create an LLM.
-    # You should pass task="classify" for classification models
+    # You should pass runner="pooling" for classification models
     llm = LLM(**vars(args))
 
     # Generate logits. The output is a list of ClassificationRequestOutputs.

examples/offline_inference/basic/embed.py

@@ -13,7 +13,7 @@ def parse_args():
     # Set example specific arguments
     parser.set_defaults(
         model="intfloat/e5-mistral-7b-instruct",
-        task="embed",
+        runner="pooling",
         enforce_eager=True,
         max_model_len=1024,
     )
@@ -30,7 +30,7 @@ def main(args: Namespace):
     ]
 
     # Create an LLM.
-    # You should pass task="embed" for embedding models
+    # You should pass runner="pooling" for embedding models
     llm = LLM(**vars(args))
 
     # Generate embedding. The output is a list of EmbeddingRequestOutputs.

examples/offline_inference/basic/score.py

@@ -12,7 +12,9 @@ def parse_args():
     parser = EngineArgs.add_cli_args(parser)
     # Set example specific arguments
     parser.set_defaults(
-        model="BAAI/bge-reranker-v2-m3", task="score", enforce_eager=True
+        model="BAAI/bge-reranker-v2-m3",
+        runner="pooling",
+        enforce_eager=True,
     )
     return parser.parse_args()
 
@@ -26,7 +28,7 @@ def main(args: Namespace):
     ]
 
     # Create an LLM.
-    # You should pass task="score" for cross-encoder models
+    # You should pass runner="pooling" for cross-encoder models
     llm = LLM(**vars(args))
 
     # Generate scores. The output is a list of ScoringRequestOutputs.

examples/offline_inference/embed_jina_embeddings_v3.py

@@ -12,7 +12,9 @@ def parse_args():
     parser = EngineArgs.add_cli_args(parser)
     # Set example specific arguments
     parser.set_defaults(
-        model="jinaai/jina-embeddings-v3", task="embed", trust_remote_code=True
+        model="jinaai/jina-embeddings-v3",
+        runner="pooling",
+        trust_remote_code=True,
     )
     return parser.parse_args()
 
@@ -29,7 +31,7 @@ def main(args: Namespace):
     ]
 
     # Create an LLM.
-    # You should pass task="embed" for embedding models
+    # You should pass runner="pooling" for embedding models
     llm = LLM(**vars(args))
 
     # Generate embedding. The output is a list of EmbeddingRequestOutputs.

examples/offline_inference/embed_matryoshka_fy.py

@@ -12,7 +12,9 @@ def parse_args():
     parser = EngineArgs.add_cli_args(parser)
     # Set example specific arguments
     parser.set_defaults(
-        model="jinaai/jina-embeddings-v3", task="embed", trust_remote_code=True
+        model="jinaai/jina-embeddings-v3",
+        runner="pooling",
+        trust_remote_code=True,
     )
     return parser.parse_args()
 
@@ -29,7 +31,7 @@ def main(args: Namespace):
     ]
 
     # Create an LLM.
-    # You should pass task="embed" for embedding models
+    # You should pass runner="pooling" for embedding models
     llm = LLM(**vars(args))
 
     # Generate embedding. The output is a list of EmbeddingRequestOutputs.

examples/offline_inference/qwen3_reranker.py

@@ -17,7 +17,7 @@ model_name = "Qwen/Qwen3-Reranker-0.6B"
 # Models converted offline using this method can not only be more efficient
 # and support the vllm score API, but also make the init parameters more
 # concise, for example.
-# llm = LLM(model="tomaarsen/Qwen3-Reranker-0.6B-seq-cls", task="score")
+# llm = LLM(model="tomaarsen/Qwen3-Reranker-0.6B-seq-cls", runner="pooling")
 
 # If you want to load the official original version, the init parameters are
 # as follows.
@@ -27,7 +27,7 @@ def get_llm() -> LLM:
     """Initializes and returns the LLM model for Qwen3-Reranker."""
     return LLM(
         model=model_name,
-        task="score",
+        runner="pooling",
         hf_overrides={
             "architectures": ["Qwen3ForSequenceClassification"],
             "classifier_from_token": ["no", "yes"],

examples/offline_inference/vision_language_pooling.py

@@ -70,7 +70,7 @@ def run_e5_v(query: Query) -> ModelRequestData:
 
     engine_args = EngineArgs(
         model="royokong/e5-v",
-        task="embed",
+        runner="pooling",
         max_model_len=4096,
         limit_mm_per_prompt={"image": 1},
     )
@@ -102,7 +102,7 @@ def run_vlm2vec(query: Query) -> ModelRequestData:
 
     engine_args = EngineArgs(
         model="TIGER-Lab/VLM2Vec-Full",
-        task="embed",
+        runner="pooling",
         max_model_len=4096,
         trust_remote_code=True,
         mm_processor_kwargs={"num_crops": 4},
@@ -122,7 +122,7 @@ def run_jinavl_reranker(query: Query) -> ModelRequestData:
 
     engine_args = EngineArgs(
         model="jinaai/jina-reranker-m0",
-        task="score",
+        runner="pooling",
         max_model_len=32768,
         trust_remote_code=True,
         mm_processor_kwargs={