[Chore] Rename model_overide_args to model_override_args (#1284)

Signed-off-by: Kai-Hsun Chen <kaihsun@anyscale.com>
Co-authored-by: Yineng Zhang <me@zhyncs.com>

python/sglang/srt/managers/controller_multi.py

@@ -71,12 +71,12 @@ class ControllerMulti:
         self,
         server_args: ServerArgs,
         port_args: PortArgs,
-        model_overide_args,
+        model_override_args,
     ):
         # Parse args
         self.server_args = server_args
         self.port_args = port_args
-        self.model_overide_args = model_overide_args
+        self.model_override_args = model_override_args
         self.load_balance_method = LoadBalanceMethod.from_str(
             server_args.load_balance_method
         )
@@ -114,7 +114,7 @@ class ControllerMulti:
                 self.server_args,
                 self.port_args,
                 pipe_controller_writer,
-                self.model_overide_args,
+                self.model_override_args,
                 True,
                 gpu_ids,
                 dp_worker_id,
@@ -189,14 +189,14 @@ def start_controller_process(
     server_args: ServerArgs,
     port_args: PortArgs,
     pipe_writer,
-    model_overide_args: dict,
+    model_override_args: dict,
 ):
     """Start a controller process."""
 
     configure_logger(server_args)
 
     try:
-        controller = ControllerMulti(server_args, port_args, model_overide_args)
+        controller = ControllerMulti(server_args, port_args, model_override_args)
     except Exception:
         pipe_writer.send(get_exception_traceback())
         raise

python/sglang/srt/managers/controller_single.py

@@ -40,7 +40,7 @@ class ControllerSingle:
         self,
         server_args: ServerArgs,
         port_args: PortArgs,
-        model_overide_args: dict,
+        model_override_args: dict,
         gpu_ids: List[int],
         is_data_parallel_worker: bool,
         dp_worker_id: int,
@@ -76,7 +76,7 @@ class ControllerSingle:
                 tp_rank_range,
                 server_args,
                 port_args.nccl_ports[dp_worker_id],
-                model_overide_args,
+                model_override_args,
             )
 
         # Launch tp rank 0
@@ -85,7 +85,7 @@ class ControllerSingle:
             0,
             server_args,
             port_args.nccl_ports[dp_worker_id],
-            model_overide_args,
+            model_override_args,
         )
         self.tp_cpu_group = self.tp_server.model_runner.tp_group.cpu_group
 
@@ -126,7 +126,7 @@ def start_controller_process(
     server_args: ServerArgs,
     port_args: PortArgs,
     pipe_writer: multiprocessing.connection.Connection,
-    model_overide_args: dict,
+    model_override_args: dict,
     is_data_parallel_worker: bool = False,
     gpu_ids: List[int] = None,
     dp_worker_id: int = None,
@@ -149,7 +149,7 @@ def start_controller_process(
         controller = ControllerSingle(
             server_args,
             port_args,
-            model_overide_args,
+            model_override_args,
             gpu_ids,
             is_data_parallel_worker,
             dp_worker_id,

python/sglang/srt/managers/tokenizer_manager.py

@@ -77,7 +77,7 @@ class TokenizerManager:
         self,
         server_args: ServerArgs,
         port_args: PortArgs,
-        model_overide_args: dict = None,
+        model_override_args: dict = None,
     ):
         self.server_args = server_args
 
@@ -95,7 +95,7 @@ class TokenizerManager:
         self.hf_config = get_config(
             self.model_path,
             trust_remote_code=server_args.trust_remote_code,
-            model_overide_args=model_overide_args,
+            model_override_args=model_override_args,
         )
         self.is_generation = is_generation_model(
             self.hf_config.architectures, self.server_args.is_embedding

python/sglang/srt/managers/tp_worker.py

@@ -76,7 +76,7 @@ class ModelTpServer:
         tp_rank: int,
         server_args: ServerArgs,
         nccl_port: int,
-        model_overide_args: dict,
+        model_override_args: dict,
     ):
         suppress_other_loggers()
 
@@ -93,7 +93,7 @@ class ModelTpServer:
             server_args.model_path,
             server_args.trust_remote_code,
             context_length=server_args.context_length,
-            model_overide_args=model_overide_args,
+            model_override_args=model_override_args,
         )
 
         self.model_runner = ModelRunner(
@@ -876,7 +876,7 @@ def run_tp_server(
     tp_rank: int,
     server_args: ServerArgs,
     nccl_port: int,
-    model_overide_args: dict,
+    model_override_args: dict,
 ):
     """Run a tensor parallel model server."""
     configure_logger(server_args, prefix=f" TP{tp_rank}")
@@ -887,7 +887,7 @@ def run_tp_server(
             tp_rank,
             server_args,
             nccl_port,
-            model_overide_args,
+            model_override_args,
         )
         tp_cpu_group = model_server.model_runner.tp_group.cpu_group
 
@@ -904,14 +904,14 @@ def launch_tp_servers(
     tp_rank_range: List[int],
     server_args: ServerArgs,
     nccl_port: int,
-    model_overide_args: dict,
+    model_override_args: dict,
 ):
     """Launch multiple tensor parallel servers."""
     procs = []
     for i in tp_rank_range:
         proc = multiprocessing.Process(
             target=run_tp_server,
-            args=(gpu_ids[i], i, server_args, nccl_port, model_overide_args),
+            args=(gpu_ids[i], i, server_args, nccl_port, model_override_args),
         )
         proc.start()
         procs.append(proc)