Reduce logs in CLI scripts and plugin loader (#18970)

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

vllm/benchmarks/datasets.py

@@ -328,9 +328,9 @@ class RandomDataset(BenchmarkDataset):
         output_high = int(output_len * (1 + range_ratio))
 
         # Add logging for debugging
-        logger.info("Sampling input_len from [%s, %s]", input_low, input_high)
-        logger.info("Sampling output_len from [%s, %s]", output_low,
-                    output_high)
+        logger.info(
+            "Sampling input_len from [%s, %s] and output_len from [%s, %s]",
+            input_low, input_high, output_low, output_high)
 
         input_lens = np.random.randint(input_low,
                                        input_high + 1,

vllm/benchmarks/latency.py

@@ -78,7 +78,6 @@ def add_cli_args(parser: argparse.ArgumentParser):
 
 
 def main(args: argparse.Namespace):
-    print(args)
     if args.profile and not envs.VLLM_TORCH_PROFILER_DIR:
         raise OSError(
             "The environment variable 'VLLM_TORCH_PROFILER_DIR' is not set. "
@@ -101,7 +100,6 @@ def main(args: argparse.Namespace):
         max_tokens=args.output_len,
         detokenize=not args.disable_detokenize,
     )
-    print(sampling_params)
     dummy_prompt_token_ids = np.random.randint(10000,
                                                size=(args.batch_size,
                                                      args.input_len))

vllm/benchmarks/throughput.py

@@ -527,7 +527,6 @@ def main(args: argparse.Namespace):
     validate_args(args)
     if args.seed is None:
         args.seed = 0
-    print(args)
     random.seed(args.seed)
     # Sample the requests.
     tokenizer = AutoTokenizer.from_pretrained(

vllm/compilation/backends.py

@@ -31,13 +31,13 @@ def make_compiler(compilation_config: CompilationConfig) -> CompilerInterface:
     if compilation_config.use_inductor:
         if envs.VLLM_USE_STANDALONE_COMPILE and is_torch_equal_or_newer(
                 "2.8.0"):
-            logger.info("Using InductorStandaloneAdaptor")
+            logger.debug("Using InductorStandaloneAdaptor")
             return InductorStandaloneAdaptor()
         else:
-            logger.info("Using InductorAdaptor")
+            logger.debug("Using InductorAdaptor")
             return InductorAdaptor()
     else:
-        logger.info("Using EagerAdaptor")
+        logger.debug("Using EagerAdaptor")
         return EagerAdaptor()
 
 

vllm/plugins/__init__.py

@@ -10,6 +10,8 @@ import vllm.envs as envs
 
 logger = logging.getLogger(__name__)
 
+DEFAULT_PLUGINS_GROUP = 'vllm.general_plugins'
+
 # make sure one process only loads plugins once
 plugins_loaded = False
 
@@ -28,19 +30,24 @@ def load_plugins_by_group(group: str) -> dict[str, Callable[[], Any]]:
         logger.debug("No plugins for group %s found.", group)
         return {}
 
-    logger.info("Available plugins for group %s:", group)
+    # Check if the only discovered plugin is the default one
+    is_default_group = (group == DEFAULT_PLUGINS_GROUP)
+    # Use INFO for non-default groups and DEBUG for the default group
+    log_level = logger.debug if is_default_group else logger.info
+
+    log_level("Available plugins for group %s:", group)
     for plugin in discovered_plugins:
-        logger.info("- %s -> %s", plugin.name, plugin.value)
+        log_level("- %s -> %s", plugin.name, plugin.value)
 
     if allowed_plugins is None:
-        logger.info("All plugins in this group will be loaded. "
-                    "Set `VLLM_PLUGINS` to control which plugins to load.")
+        log_level("All plugins in this group will be loaded. "
+                  "Set `VLLM_PLUGINS` to control which plugins to load.")
 
     plugins = dict[str, Callable[[], Any]]()
     for plugin in discovered_plugins:
         if allowed_plugins is None or plugin.name in allowed_plugins:
             if allowed_plugins is not None:
-                logger.info("Loading plugin %s", plugin.name)
+                log_level("Loading plugin %s", plugin.name)
 
             try:
                 func = plugin.load()
@@ -80,7 +87,7 @@ def load_general_plugins():
             # see https://docs.habana.ai/en/latest/PyTorch/Inference_on_PyTorch/Inference_Using_HPU_Graphs.html # noqa: E501
             os.environ['PT_HPU_ENABLE_LAZY_COLLECTIVES'] = 'true'
 
-    plugins = load_plugins_by_group(group='vllm.general_plugins')
+    plugins = load_plugins_by_group(group=DEFAULT_PLUGINS_GROUP)
     # general plugins, we only need to execute the loaded functions
     for func in plugins.values():
         func()