Update LoRA handling for Qwen2-VL to conform to Transformers

Author: DarkLight1337

requirements-common.txt

@@ -4,7 +4,7 @@ numpy < 2.0.0
 requests >= 2.26.0
 tqdm
 py-cpuinfo
-transformers >= 4.45.0  # Required for Llama 3.2.
+transformers >= 4.45.2  # Required for Llama 3.2 and Qwen2-VL.
 tokenizers >= 0.19.1  # Required for Llama 3.
 protobuf # Required by LlamaTokenizer.
 fastapi >= 0.107.0, < 0.113.0; python_version < '3.9'

vllm/config.py

@@ -1723,16 +1723,25 @@ def _get_and_verify_max_len(
 
     rope_scaling = getattr(hf_config, "rope_scaling", None)
     if rope_scaling is not None:
+        # Backwards compatibility. Although HF prefers "rope_type", we still
+        # have code that accesses "type"
         if "type" in rope_scaling:
-            rope_type = rope_scaling["type"]
+            rope_type = rope_scaling["rope_type"] = rope_scaling["type"]
         elif "rope_type" in rope_scaling:
-            rope_type = rope_scaling["rope_type"]
+            rope_type = rope_scaling["type"] = rope_scaling["rope_type"]
         else:
             raise ValueError(
                 "rope_scaling must have a 'type' or 'rope_type' key.")
 
-        # The correct one should be "longrope", kept "su" here
-        # to be backward compatible
+        # Backwards compatibility
+        if rope_type == "su":
+            rope_scaling["rope_type"] = rope_type = "longrope"
+            logger.warning("Replacing legacy rope_type 'su' with 'longrope'")
+        elif rope_type == "mrope":
+            assert "mrope_section" in rope_scaling
+            rope_scaling["rope_type"] = rope_type = "default"
+            logger.warning("Replacing legacy rope_type 'mrope' with 'default'")
+
         if rope_type not in ("su", "longrope", "llama3"):
             if disable_sliding_window:
                 # TODO(robertgshaw): Find a model that supports rope_scaling
@@ -1742,11 +1751,10 @@ def _get_and_verify_max_len(
                     "with rope_scaling. Please raise an issue so we can "
                     "investigate.")
 
-            if rope_type == "mrope":
-                scaling_factor = 1
-            else:
-                assert "factor" in rope_scaling
-                scaling_factor = rope_scaling["factor"]
+            # NOTE: rope_type == "default" does not define factor
+            # https://github.com/huggingface/transformers/blob/v4.45.2/src/transformers/modeling_rope_utils.py
+            scaling_factor = rope_scaling.get("factor", 1.0)
+
             if rope_type == "yarn":
                 derived_max_model_len = rope_scaling[
                     "original_max_position_embeddings"]

vllm/model_executor/layers/rotary_embedding.py

@@ -922,11 +922,9 @@ def get_rope(
     else:
         scaling_type = rope_scaling[
             "type"] if "type" in rope_scaling else rope_scaling["rope_type"]
-        # The correct one should be "longrope" but keep "su" here
-        # for backward compatible
-        if scaling_type not in {"su", "longrope"}:
-            scaling_factor = rope_scaling.get("factor", 1.0)
+
         if scaling_type == "llama3":
+            scaling_factor = rope_scaling["factor"]
             low_freq_factor = rope_scaling["low_freq_factor"]
             high_freq_factor = rope_scaling["high_freq_factor"]
             original_max_position = rope_scaling[
@@ -937,16 +935,39 @@ def get_rope(
                                                scaling_factor, low_freq_factor,
                                                high_freq_factor,
                                                original_max_position)
+        elif scaling_type == "default":
+            if "mrope_section" in rope_scaling:
+                rotary_emb = MRotaryEmbedding(
+                    head_size,
+                    rotary_dim,
+                    max_position,
+                    base,
+                    is_neox_style,
+                    dtype,
+                    mrope_section=rope_scaling["mrope_section"],
+                )
+            else:
+                rotary_emb = RotaryEmbedding(
+                    head_size,
+                    rotary_dim,
+                    max_position,
+                    base,
+                    is_neox_style,
+                    dtype,
+                )
         elif scaling_type == "linear":
+            scaling_factor = rope_scaling["factor"]
             rotary_emb = LinearScalingRotaryEmbedding(head_size, rotary_dim,
                                                       max_position, base,
                                                       is_neox_style,
                                                       scaling_factor, dtype)
         elif scaling_type == "dynamic":
+            scaling_factor = rope_scaling["factor"]
             rotary_emb = DynamicNTKScalingRotaryEmbedding(
                 head_size, rotary_dim, max_position, base, is_neox_style,
                 scaling_factor, dtype)
         elif scaling_type == "yarn":
+            scaling_factor = rope_scaling["factor"]
             original_max_position = rope_scaling[
                 "original_max_position_embeddings"]
             extra_kwargs = {
@@ -961,6 +982,7 @@ def get_rope(
                                                     scaling_factor, dtype,
                                                     **extra_kwargs)
         elif scaling_type == "deepseek_yarn":
+            scaling_factor = rope_scaling["factor"]
             original_max_position = rope_scaling[
                 "original_max_position_embeddings"]
             # assert max_position == original_max_position * scaling_factor
@@ -973,9 +995,7 @@ def get_rope(
             rotary_emb = DeepseekScalingRotaryEmbedding(
                 head_size, rotary_dim, original_max_position, base,
                 is_neox_style, scaling_factor, dtype, **extra_kwargs)
-        # The correct one should be "longrope" but keep "su" here
-        # for backward compatible
-        elif scaling_type == "su" or scaling_type == "longrope":
+        elif scaling_type == "longrope":
             short_factor = rope_scaling["short_factor"]
             long_factor = rope_scaling["long_factor"]
             original_max_position = rope_scaling[
@@ -989,16 +1009,6 @@ def get_rope(
                 head_size, rotary_dim, max_position, original_max_position,
                 base, is_neox_style, dtype, short_factor, long_factor,
                 **extra_kwargs)
-        elif scaling_type == "mrope":
-            rotary_emb = MRotaryEmbedding(
-                head_size,
-                rotary_dim,
-                max_position,
-                base,
-                is_neox_style,
-                dtype,
-                mrope_section=rope_scaling["mrope_section"],
-            )
         else:
             raise ValueError(f"Unknown RoPE scaling type {scaling_type}")
     _ROPE_DICT[key] = rotary_emb

vllm/model_executor/models/qwen2_vl.py

@@ -34,6 +34,8 @@
 from transformers.image_utils import (get_image_size,
                                       infer_channel_dimension_format,
                                       to_numpy_array)
+from transformers.models.qwen2_vl.configuration_qwen2_vl import (
+    Qwen2VLConfig, Qwen2VLVisionConfig)
 from transformers.models.qwen2_vl.image_processing_qwen2_vl import (
     make_batched_images, make_batched_videos, smart_resize)
 
@@ -62,8 +64,6 @@
 from vllm.multimodal.image import cached_get_image_processor
 from vllm.platforms import current_platform
 from vllm.sequence import IntermediateTensors, SequenceData
-from vllm.transformers_utils.configs.qwen2vl import (Qwen2VLConfig,
-                                                     Qwen2VLVisionConfig)
 from vllm.transformers_utils.processor import get_processor
 from vllm.utils import is_cpu
 

vllm/transformers_utils/config.py

@@ -23,8 +23,8 @@
                                              MedusaConfig, MllamaConfig,
                                              MLPSpeculatorConfig, MPTConfig,
                                              NemotronConfig, NVLM_D_Config,
-                                             Qwen2VLConfig, RWConfig,
-                                             SolarConfig, UltravoxConfig)
+                                             RWConfig, SolarConfig,
+                                             UltravoxConfig)
 # yapf: enable
 from vllm.transformers_utils.utils import check_gguf_file
 
@@ -57,7 +57,6 @@
     "NVLM_D": NVLM_D_Config,
     "solar": SolarConfig,
     "ultravox": UltravoxConfig,
-    "qwen2_vl": Qwen2VLConfig,
     **_CONFIG_REGISTRY_OVERRIDE_HF
 }
 

vllm/transformers_utils/configs/__init__.py

@@ -14,8 +14,6 @@
 from vllm.transformers_utils.configs.mpt import MPTConfig
 from vllm.transformers_utils.configs.nemotron import NemotronConfig
 from vllm.transformers_utils.configs.nvlm_d import NVLM_D_Config
-from vllm.transformers_utils.configs.qwen2vl import (Qwen2VLConfig,
-                                                     Qwen2VLVisionConfig)
 from vllm.transformers_utils.configs.solar import SolarConfig
 from vllm.transformers_utils.configs.ultravox import UltravoxConfig
 
@@ -35,6 +33,4 @@
     "NVLM_D_Config",
     "SolarConfig",
     "UltravoxConfig",
-    "Qwen2VLConfig",
-    "Qwen2VLVisionConfig",
 ]