[Pixtral] Improve loading (#11040)

vllm/model_executor/models/pixtral.py

@@ -1,6 +1,5 @@
 from dataclasses import dataclass, fields
 from functools import cached_property
-from itertools import tee
 from typing import Iterable, List, Mapping, Optional, Set, Tuple, Union
 
 import numpy
@@ -359,38 +358,33 @@ class PixtralForConditionalGeneration(nn.Module, SupportsMultiModal,
         def is_vision_lang_adapter_weights(weight: Tuple[str, torch.Tensor]):
             return weight[0].startswith("vision_language_adapter")
 
-        def is_vision_weights(weight: Tuple[str, torch.Tensor]):
+        # Get references to parameters for direct loading
-            return is_vision_encoder_weights(
-                weight) or is_vision_lang_adapter_weights(weight)
-
-        llm_weights, vision_encoder_weights, vision_lang_adapter_weights = tee(
-            weights, 3)
-
-        # llm
-        llm_weights = filter(lambda x: not is_vision_weights(x), llm_weights)
-        self.language_model.load_weights(llm_weights)
-
-        # vision encoder
-        vision_encoder_weights = filter(is_vision_encoder_weights,
-                                        vision_encoder_weights)
         vision_encoder_dict = dict(self.vision_encoder.named_parameters())
-        for name, loaded_weight in vision_encoder_weights:
+        vision_lang_adapter_dict = dict(
-            # cut 'vision_encoder.'
-            name = '.'.join(name.split(".")[1:])
-            param = vision_encoder_dict[name]
-
-            default_weight_loader(param, loaded_weight)
-
-        # adapter
-        vision_lang_adapter_weights = filter(is_vision_lang_adapter_weights,
-                                             vision_lang_adapter_weights)
-        vision_lang_adpter_dict = dict(
             self.vision_language_adapter.named_parameters())
-        for name, loaded_weight in vision_lang_adapter_weights:
+
-            # cut 'vision_language_adapter.'
+        def llm_weights_generator():
-            name = '.'.join(name.split(".")[1:])
+            # Single pass over weights
-            param = vision_lang_adpter_dict[name]
+            for name, w in weights:
-            default_weight_loader(param, loaded_weight)
+                if is_vision_encoder_weights((name, w)):
+                    # Load vision encoder weights directly
+                    trimmed_name = '.'.join(name.split(".")[1:])
+                    param = vision_encoder_dict[trimmed_name]
+                    with torch.no_grad():
+                        default_weight_loader(param, w)
+                elif is_vision_lang_adapter_weights((name, w)):
+                    # Load vision-language adapter weights directly
+                    trimmed_name = '.'.join(name.split(".")[1:])
+                    param = vision_lang_adapter_dict[trimmed_name]
+                    with torch.no_grad():
+                        default_weight_loader(param, w)
+                else:
+                    # LLM weights: yield them to be loaded
+                    # by language_model.load_weights
+                    yield (name, w)
+
+        # Now we call the language model load with the generator
+        self.language_model.load_weights(llm_weights_generator())
 
 
 # Vision encoder