[Cherry-Pick][Others]update paddleformer 1.0.0 PaddlePaddle#6496 (PaddlePaddle#6752)

bukejiyu · web-flow · commit 69804c901357 · 2026-03-27T01:24:22.000-07:00
diff --git a/fastdeploy/input/ernie4_5_processor.py b/fastdeploy/input/ernie4_5_processor.py
@@ -68,7 +68,10 @@ def __init__(self, model_name_or_path, reasoning_parser_obj=None, tool_parser_ob
                                    {self.tokenizer.bos_token_id}, \
                                    eos_token is {self.tokenizer.eos_token}, {self.tokenizer.eos_token_id} "
         )
-        from paddleformers.trl.llm_utils import get_eos_token_id
+        try:
+            from paddleformers.trl.llm_utils import get_eos_token_id
+        except Exception:
+            from paddleformers.cli.utils.llm_utils import get_eos_token_id
 
         self.eos_token_ids = get_eos_token_id(self.tokenizer, self.generation_config)
         self.eos_token_id_len = len(self.eos_token_ids)
diff --git a/fastdeploy/input/ernie4_5_tokenizer.py b/fastdeploy/input/ernie4_5_tokenizer.py
@@ -23,7 +23,10 @@
 import paddle
 import sentencepiece as spm
 from paddleformers.transformers import PretrainedTokenizer
-from paddleformers.transformers.tokenizer_utils_base import PaddingStrategy, TextInput
+from paddleformers.transformers.legacy.tokenizer_utils_base import (
+    PaddingStrategy,
+    TextInput,
+)
 from paddleformers.utils.log import logger
 
 
diff --git a/fastdeploy/input/ernie4_5_vl_processor/ernie4_5_vl_processor.py b/fastdeploy/input/ernie4_5_vl_processor/ernie4_5_vl_processor.py
@@ -72,7 +72,10 @@ def __init__(
             self.generation_config = None
 
         # self.eos_token_ids = [self.tokenizer.eos_token_id]
-        from paddleformers.trl.llm_utils import get_eos_token_id
+        try:
+            from paddleformers.trl.llm_utils import get_eos_token_id
+        except Exception:
+            from paddleformers.cli.utils.llm_utils import get_eos_token_id
 
         self.eos_token_ids = get_eos_token_id(self.tokenizer, self.generation_config)
         self.eos_token_id_len = len(self.eos_token_ids)
diff --git a/fastdeploy/input/ernie4_5_vl_processor/image_preprocessor/image_preprocessor_adaptive.py b/fastdeploy/input/ernie4_5_vl_processor/image_preprocessor/image_preprocessor_adaptive.py
@@ -42,7 +42,7 @@
     to_numpy_array,
     valid_images,
 )
-from paddleformers.transformers.tokenizer_utils_base import TensorType
+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType
 from PIL import Image
 
 from fastdeploy.utils import data_processor_logger
diff --git a/fastdeploy/input/qwen3_vl_processor/image_processor.py b/fastdeploy/input/qwen3_vl_processor/image_processor.py
@@ -38,7 +38,7 @@
     to_numpy_array,
     valid_images,
 )
-from paddleformers.transformers.tokenizer_utils_base import TensorType
+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType
 from PIL import Image
 
 from fastdeploy.utils import data_processor_logger
diff --git a/fastdeploy/input/qwen_vl_processor/image_processor.py b/fastdeploy/input/qwen_vl_processor/image_processor.py
@@ -38,7 +38,7 @@
     to_numpy_array,
     valid_images,
 )
-from paddleformers.transformers.tokenizer_utils_base import TensorType
+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType
 from PIL import Image
 
 from fastdeploy.utils import data_processor_logger
diff --git a/fastdeploy/input/text_processor.py b/fastdeploy/input/text_processor.py
@@ -194,7 +194,10 @@ def __init__(self, model_name_or_path, reasoning_parser_obj=None, tool_parser_ob
                                 eos_token is {self.tokenizer.eos_token}, {self.tokenizer.eos_token_id} "
         )
 
-        from paddleformers.trl.llm_utils import get_eos_token_id
+        try:
+            from paddleformers.trl.llm_utils import get_eos_token_id
+        except Exception:
+            from paddleformers.cli.utils.llm_utils import get_eos_token_id
 
         self.eos_token_ids = get_eos_token_id(self.tokenizer, self.generation_config)
         data_processor_logger.info(
diff --git a/fastdeploy/input/v1/ernie4_5_processor.py b/fastdeploy/input/v1/ernie4_5_processor.py
@@ -68,7 +68,10 @@ def __init__(self, model_name_or_path, reasoning_parser_obj=None, tool_parser_ob
                                    {self.tokenizer.bos_token_id}, \
                                    eos_token is {self.tokenizer.eos_token}, {self.tokenizer.eos_token_id} "
         )
-        from paddleformers.trl.llm_utils import get_eos_token_id
+        try:
+            from paddleformers.trl.llm_utils import get_eos_token_id
+        except Exception:
+            from paddleformers.cli.utils.llm_utils import get_eos_token_id
 
         self.eos_token_ids = get_eos_token_id(self.tokenizer, self.generation_config)
         self.eos_token_id_len = len(self.eos_token_ids)
diff --git a/fastdeploy/input/v1/ernie4_5_vl_processor/ernie4_5_vl_processor.py b/fastdeploy/input/v1/ernie4_5_vl_processor/ernie4_5_vl_processor.py
@@ -71,7 +71,10 @@ def __init__(
             self.generation_config = None
 
         # self.eos_token_ids = [self.tokenizer.eos_token_id]
-        from paddleformers.trl.llm_utils import get_eos_token_id
+        try:
+            from paddleformers.trl.llm_utils import get_eos_token_id
+        except Exception:
+            from paddleformers.cli.utils.llm_utils import get_eos_token_id
 
         self.eos_token_ids = get_eos_token_id(self.tokenizer, self.generation_config)
         self.eos_token_id_len = len(self.eos_token_ids)
diff --git a/fastdeploy/input/v1/ernie4_5_vl_processor/image_preprocessor/image_preprocessor_adaptive.py b/fastdeploy/input/v1/ernie4_5_vl_processor/image_preprocessor/image_preprocessor_adaptive.py
@@ -42,7 +42,7 @@
     to_numpy_array,
     valid_images,
 )
-from paddleformers.transformers.tokenizer_utils_base import TensorType
+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType
 from PIL import Image
 
 from fastdeploy.utils import data_processor_logger
diff --git a/fastdeploy/input/v1/qwen3_vl_processor/image_processor.py b/fastdeploy/input/v1/qwen3_vl_processor/image_processor.py
@@ -38,7 +38,7 @@
     to_numpy_array,
     valid_images,
 )
-from paddleformers.transformers.tokenizer_utils_base import TensorType
+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType
 from PIL import Image
 
 from fastdeploy.utils import data_processor_logger
diff --git a/fastdeploy/input/v1/qwen_vl_processor/image_processor.py b/fastdeploy/input/v1/qwen_vl_processor/image_processor.py
@@ -38,7 +38,7 @@
     to_numpy_array,
     valid_images,
 )
-from paddleformers.transformers.tokenizer_utils_base import TensorType
+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType
 from PIL import Image
 
 from fastdeploy.utils import data_processor_logger
diff --git a/fastdeploy/input/v1/text_processor.py b/fastdeploy/input/v1/text_processor.py
@@ -180,7 +180,10 @@ def __init__(self, model_name_or_path, reasoning_parser_obj=None, tool_parser_ob
                                 eos_token is {self.tokenizer.eos_token}, {self.tokenizer.eos_token_id} "
         )
 
-        from paddleformers.trl.llm_utils import get_eos_token_id
+        try:
+            from paddleformers.trl.llm_utils import get_eos_token_id
+        except Exception:
+            from paddleformers.cli.utils.llm_utils import get_eos_token_id
 
         self.eos_token_ids = get_eos_token_id(self.tokenizer, self.generation_config)
         data_processor_logger.info(
diff --git a/fastdeploy/model_executor/guided_decoding/xgrammar_backend.py b/fastdeploy/model_executor/guided_decoding/xgrammar_backend.py
@@ -22,6 +22,7 @@
 import paddle
 import torch
 
+paddle.compat.disable_torch_proxy()
 from fastdeploy.config import FDConfig
 from fastdeploy.engine.request import Request
 from fastdeploy.model_executor.guided_decoding import (
diff --git a/fastdeploy/model_executor/models/ernie4_5_vl/dfnrope/modeling.py b/fastdeploy/model_executor/models/ernie4_5_vl/dfnrope/modeling.py
@@ -26,7 +26,6 @@
     ColumnParallelLinear,
     RowParallelLinear,
 )
-from paddle.distributed.fleet.utils import recompute
 from paddle.nn.functional.flash_attention import (
     flash_attn_unpadded as flash_attn_varlen_func,
 )
@@ -638,17 +637,12 @@ def forward(self, hidden_states: paddle.Tensor, grid_thw: paddle.Tensor, num_pad
         else:
             cu_seqlens = F.pad(cu_seqlens, (1, 0), value=0)
 
-        vit_num_recompute_layers = getattr(self.config, "vit_num_recompute_layers", self.config.depth)
-
         for idx, blk in enumerate(self.blocks):
-            if self.config.recompute and self.training and idx < vit_num_recompute_layers:
-                hidden_states = recompute(blk, hidden_states, cu_seqlens, rotary_pos_emb)
-            else:
-                hidden_states = blk(
-                    hidden_states,
-                    cu_seqlens=cu_seqlens,
-                    rotary_pos_emb=rotary_pos_emb,
-                )
+            hidden_states = blk(
+                hidden_states,
+                cu_seqlens=cu_seqlens,
+                rotary_pos_emb=rotary_pos_emb,
+            )
 
         # ret = self.merger(hidden_states)
         # ret = hidden_states
diff --git a/requirements.txt b/requirements.txt
@@ -10,7 +10,7 @@ tqdm
 pynvml
 uvicorn>=0.38.0
 fastapi
-paddleformers @ https://paddle-qa.bj.bcebos.com/ernie/paddleformers-0.4.0.post20251222-py3-none-any.whl
+paddleformers==1.0.0
 redis
 etcd3
 httpx
@@ -49,3 +49,4 @@ py-cpuinfo
 flashinfer-python-paddle
 flash_mask @ https://paddle-qa.bj.bcebos.com/ernie/flash_mask-4.0.post20260128-py3-none-any.whl
 arctic_inference @ https://paddle-qa.bj.bcebos.com/ernie/arctic_inference-0.1.3-cp310-cp310-linux_x86_64.whl
+transformers>=4.55.1,<5.0.0
diff --git a/requirements_dcu.txt b/requirements_dcu.txt
@@ -10,7 +10,7 @@ tqdm
 pynvml
 uvicorn>=0.38.0
 fastapi
-paddleformers @ https://paddle-qa.bj.bcebos.com/ernie/paddleformers-0.4.0.post20251222-py3-none-any.whl
+paddleformers==1.0.0
 redis
 etcd3
 httpx
@@ -39,3 +39,4 @@ opentelemetry-instrumentation-fastapi
 opentelemetry-instrumentation-logging>=0.57b0
 partial_json_parser
 py-cpuinfo
+transformers>=4.55.1,<5.0.0
diff --git a/requirements_iluvatar.txt b/requirements_iluvatar.txt
@@ -10,7 +10,7 @@ tqdm
 pynvml
 uvicorn>=0.38.0
 fastapi
-paddleformers @ https://paddle-qa.bj.bcebos.com/ernie/paddleformers-0.4.0.post20251222-py3-none-any.whl
+paddleformers==1.0.0
 redis
 etcd3
 httpx
@@ -41,3 +41,4 @@ partial_json_parser
 msgspec
 safetensors>=0.7.0
 py-cpuinfo
+transformers>=4.55.1,<5.0.0
diff --git a/requirements_metaxgpu.txt b/requirements_metaxgpu.txt
@@ -10,7 +10,7 @@ tqdm
 pynvml
 uvicorn>=0.38.0
 fastapi
-paddleformers==0.4.1
+paddleformers==1.0.0
 redis
 etcd3
 httpx
@@ -46,3 +46,4 @@ einops
 setproctitle
 aistudio_sdk
 py-cpuinfo
+transformers>=4.55.1,<5.0.0
diff --git a/tests/ci_use/utils/rollout_model.py b/tests/ci_use/utils/rollout_model.py
@@ -14,7 +14,10 @@
 
 import argparse
 
-from paddleformers.trl.llm_utils import init_dist_env
+try:
+    from paddleformers.trl.llm_utils import init_dist_env
+except Exception:
+    from paddleformers.cli.utils.llm_utils import init_dist_env
 
 from fastdeploy.rl.rollout_config import RolloutModelConfig
 from fastdeploy.rl.rollout_model import RolloutModel
diff --git a/tests/e2e/utils/rollout_model.py b/tests/e2e/utils/rollout_model.py
@@ -14,7 +14,10 @@
 
 import argparse
 
-from paddleformers.trl.llm_utils import init_dist_env
+try:
+    from paddleformers.trl.llm_utils import init_dist_env
+except Exception:
+    from paddleformers.cli.utils.llm_utils import init_dist_env
 
 from fastdeploy.rl.rollout_config import RolloutModelConfig
 from fastdeploy.rl.rollout_model import RolloutModel
diff --git a/tests/input/test_ernie4_5_processor.py b/tests/input/test_ernie4_5_processor.py
@@ -143,7 +143,7 @@ def setUp(self):
             f"{MODULE_PATH}.Ernie4_5Tokenizer.from_pretrained", side_effect=lambda path: MockTokenizer()
         )
         self.eos_patcher = patch(
-            "paddleformers.trl.llm_utils.get_eos_token_id",
+            "paddleformers.cli.utils.llm_utils.get_eos_token_id",
             side_effect=lambda tokenizer, cfg: [tokenizer.eos_token_id],
         )
 
diff --git a/tests/input/test_ernie_vl_processor.py b/tests/input/test_ernie_vl_processor.py
@@ -147,7 +147,7 @@ def test_init(self):
                     "fastdeploy.input.ernie4_5_vl_processor.ernie4_5_vl_processor.GenerationConfig"
                 ) as mock_gen_config:
                     mock_gen_config.from_pretrained.return_value = MagicMock()
-                    with patch("paddleformers.trl.llm_utils.get_eos_token_id") as mock_get_eos:
+                    with patch("paddleformers.cli.utils.llm_utils.get_eos_token_id") as mock_get_eos:
                         mock_get_eos.return_value = [1, 2]
 
                         # Test normal initialization
diff --git a/tests/input/test_text_processor.py b/tests/input/test_text_processor.py
@@ -159,7 +159,7 @@ def from_pretrained(cls, *args, **kwargs):
     hf_transformers_module = types.ModuleType("transformers")
     hf_transformers_module.AutoTokenizer = DummyHFTokenizer
 
-    llm_utils_module = types.ModuleType("paddleformers.trl.llm_utils")
+    llm_utils_module = types.ModuleType("paddleformers.cli.utils.llm_utils")
     llm_utils_module.get_eos_token_id = lambda tokenizer, config: [tokenizer.eos_token_id]
 
     fastdeploy_module = types.ModuleType("fastdeploy")
@@ -174,7 +174,7 @@ def from_pretrained(cls, *args, **kwargs):
         "paddleformers.generation": generation_module,
         "paddleformers.transformers": transformers_module,
         "transformers": hf_transformers_module,
-        "paddleformers.trl.llm_utils": llm_utils_module,
+        "paddleformers.cli.utils.llm_utils": llm_utils_module,
     }
 
 
diff --git a/tests/input/v1/test_ernie4_5_processor.py b/tests/input/v1/test_ernie4_5_processor.py
@@ -154,7 +154,7 @@ def setUp(self):
             f"{MODULE_PATH}.Ernie4_5Tokenizer.from_pretrained", side_effect=lambda path: MockTokenizer()
         )
         self.eos_patcher = patch(
-            "paddleformers.trl.llm_utils.get_eos_token_id",
+            "paddleformers.cli.utils.llm_utils.get_eos_token_id",
             side_effect=lambda tokenizer, cfg: [tokenizer.eos_token_id],
         )
 
diff --git a/tests/input/v1/test_ernie_vl_processor.py b/tests/input/v1/test_ernie_vl_processor.py
@@ -153,7 +153,7 @@ def test_init(self):
                     "fastdeploy.input.v1.ernie4_5_vl_processor.ernie4_5_vl_processor.GenerationConfig"
                 ) as mock_gen_config:
                     mock_gen_config.from_pretrained.return_value = MagicMock()
-                    with patch("paddleformers.trl.llm_utils.get_eos_token_id") as mock_get_eos:
+                    with patch("paddleformers.cli.utils.llm_utils.get_eos_token_id") as mock_get_eos:
                         mock_get_eos.return_value = [1, 2]
 
                         # Test normal initialization
diff --git a/tests/input/v1/test_text_processor.py b/tests/input/v1/test_text_processor.py
@@ -155,7 +155,7 @@ def from_pretrained(cls, *args, **kwargs):
     hf_transformers_module = types.ModuleType("transformers")
     hf_transformers_module.AutoTokenizer = DummyHFTokenizer
 
-    llm_utils_module = types.ModuleType("paddleformers.trl.llm_utils")
+    llm_utils_module = types.ModuleType("paddleformers.cli.utils.llm_utils")
     llm_utils_module.get_eos_token_id = lambda tokenizer, config: [tokenizer.eos_token_id]
 
     fastdeploy_module = types.ModuleType("fastdeploy")
@@ -170,7 +170,7 @@ def from_pretrained(cls, *args, **kwargs):
         "paddleformers.generation": generation_module,
         "paddleformers.transformers": transformers_module,
         "transformers": hf_transformers_module,
-        "paddleformers.trl.llm_utils": llm_utils_module,
+        "paddleformers.cli.utils.llm_utils": llm_utils_module,
     }
 
 
diff --git a/tests/layers/test_guided_decoding.py b/tests/layers/test_guided_decoding.py
@@ -25,6 +25,8 @@
 
 import paddle
 
+import fastdeploy  # noqa: F401
+
 mock_torch = MagicMock()
 mock_xgrammar = MagicMock()
 sys.modules["torch"] = mock_torch
diff --git a/tests/layers/test_sampler.py b/tests/layers/test_sampler.py
@@ -20,6 +20,7 @@
 import paddle
 import paddle.nn.functional as F
 
+import fastdeploy  # noqa: F401
 from fastdeploy.config import (
     CacheConfig,
     FDConfig,
diff --git a/tests/model_executor/guided_decoding/test_guidance_checker.py b/tests/model_executor/guided_decoding/test_guidance_checker.py
@@ -21,6 +21,8 @@
 
 import pytest
 
+import fastdeploy  # noqa: F401
+
 # Check if llguidance can be imported
 HAS_LLGUIDANCE = False
 try:
diff --git a/tests/model_executor/guided_decoding/test_xgrammar_checker.py b/tests/model_executor/guided_decoding/test_xgrammar_checker.py
@@ -17,6 +17,8 @@
 import unittest
 from unittest.mock import MagicMock
 
+import fastdeploy  # noqa: F401
+
 mock_torch = MagicMock()
 mock_xgrammar = MagicMock()
 

Original file line number	Diff line number	Diff line change
`@@ -42,7 +42,7 @@`
`42`	`42`	`to_numpy_array,`
`43`	`43`	`valid_images,`
`44`	`44`	`)`
`45`		`-from paddleformers.transformers.tokenizer_utils_base import TensorType`
	`45`	`+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType`
`46`	`46`	`from PIL import Image`
`47`	`47`
`48`	`48`	`from fastdeploy.utils import data_processor_logger`
Original file line number	Diff line number	Diff line change
`@@ -38,7 +38,7 @@`
`38`	`38`	`to_numpy_array,`
`39`	`39`	`valid_images,`
`40`	`40`	`)`
`41`		`-from paddleformers.transformers.tokenizer_utils_base import TensorType`
	`41`	`+from paddleformers.transformers.legacy.tokenizer_utils_base import TensorType`
`42`	`42`	`from PIL import Image`
`43`	`43`
`44`	`44`	`from fastdeploy.utils import data_processor_logger`