[V0 deprecation] Remove more V0 references (#29088)

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

examples/offline_inference/reproducibility.py

@@ -11,13 +11,9 @@ import random
 
 from vllm import LLM, SamplingParams
 
-# V1 only: Turn off multiprocessing to make the scheduling deterministic.
+# Turn off multiprocessing to make the scheduling deterministic.
 os.environ["VLLM_ENABLE_V1_MULTIPROCESSING"] = "0"
 
-# V0 only: Set the global seed. The default seed is None, which is
-# not reproducible.
-SEED = 42
-
 prompts = [
     "Hello, my name is",
     "The president of the United States is",
@@ -28,7 +24,7 @@ sampling_params = SamplingParams(temperature=0.8, top_p=0.95)
 
 
 def main():
-    llm = LLM(model="facebook/opt-125m", seed=SEED)
+    llm = LLM(model="facebook/opt-125m")
     outputs = llm.generate(prompts, sampling_params)
     print("-" * 50)
     for output in outputs:

examples/offline_inference/rlhf_utils.py

@@ -30,8 +30,8 @@ class WorkerExtension:
     """
     The class for vLLM's worker to inherit from.
     By defining an extension class, the code can work no matter what is
-    the underlying worker class. This way, the code can be compatible
-    with both vLLM V0 and V1.
+    the underlying worker class.
+
     NOTE: we define this class in a separate module, and the main module
     should pass the full qualified name as `worker_extension_cls` argument.
     """
@@ -96,8 +96,8 @@ class ColocateWorkerExtension:
     """
     The class for vLLM's worker to inherit from, in the colocate setting.
     By defining an extension class, the code can work no matter what is
-    the underlying worker class. This way, the code can be compatible
-    with both vLLM V0 and V1.
+    the underlying worker class.
+
     NOTE: we define this class in a separate module, and the main module
     should pass the full qualified name as `worker_extension_cls` argument.
     """

examples/offline_inference/save_sharded_state.py

@@ -67,22 +67,9 @@ def main(args):
     Path(args.output).mkdir(exist_ok=True)
     # Dump worker states to output directory
 
-    # Check which engine version is being used
-    is_v1_engine = hasattr(llm.llm_engine, "engine_core")
-
-    if is_v1_engine:
-        # For V1 engine, we need to use engine_core.save_sharded_state
-        print("Using V1 engine save path")
-        llm.llm_engine.engine_core.save_sharded_state(
-            path=args.output, pattern=args.file_pattern, max_size=args.max_file_size
-        )
-    else:
-        # For V0 engine
-        print("Using V0 engine save path")
-        model_executor = llm.llm_engine.model_executor
-        model_executor.save_sharded_state(
-            path=args.output, pattern=args.file_pattern, max_size=args.max_file_size
-        )
+    llm.llm_engine.engine_core.save_sharded_state(
+        path=args.output, pattern=args.file_pattern, max_size=args.max_file_size
+    )
 
     # Copy metadata files to output directory
     for file in os.listdir(model_path):

examples/offline_inference/spec_decode.py

@@ -158,11 +158,7 @@ def main(args):
             print(f"generated text: {output.outputs[0].text}")
             print("-" * 50)
 
-    try:
-        metrics = llm.get_metrics()
-    except AssertionError:
-        print("Metrics are not supported in the V0 engine.")
-        return
+    metrics = llm.get_metrics()
 
     total_num_output_tokens = sum(
         len(output.outputs[0].token_ids) for output in outputs