Merge branch 'master' into lumina

backend/python/autogptq/requirements.txt

@@ -1,6 +1,6 @@
 accelerate
 auto-gptq==0.7.1
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 certifi
 transformers

backend/python/bark/requirements.txt

@@ -1,4 +1,4 @@
 bark==0.1.5
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 certifi

backend/python/common/template/requirements.txt

@@ -1,3 +1,3 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 grpcio-tools

backend/python/coqui/requirements-cpu.txt

@@ -1,4 +1,4 @@
-transformers
+transformers==4.48.3
 accelerate
 torch==2.4.1
 coqui-tts

backend/python/coqui/requirements-cublas11.txt

@@ -1,6 +1,6 @@
 --extra-index-url https://download.pytorch.org/whl/cu118
 torch==2.4.1+cu118
 torchaudio==2.4.1+cu118
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/coqui/requirements-cublas12.txt

@@ -1,5 +1,5 @@
 torch==2.4.1
 torchaudio==2.4.1
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/coqui/requirements-hipblas.txt

@@ -1,6 +1,6 @@
 --extra-index-url https://download.pytorch.org/whl/rocm6.0
 torch==2.4.1+rocm6.0
 torchaudio==2.4.1+rocm6.0
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/coqui/requirements-intel.txt

@@ -5,6 +5,6 @@ torchaudio==2.3.1+cxx11.abi
 oneccl_bind_pt==2.3.100+xpu
 optimum[openvino]
 setuptools
-transformers
+transformers==4.48.3
 accelerate
 coqui-tts

backend/python/coqui/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 certifi
 packaging==24.1

backend/python/diffusers/backend.py

@@ -159,6 +159,18 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
                 torchType = torch.float16
                 variant = "fp16"
 
+            options = request.Options
+
+            # empty dict
+            self.options = {}
+
+            # The options are a list of strings in this form optname:optvalue
+            # We are storing all the options in a dict so we can use it later when
+            # generating the images
+            for opt in options:
+                key, value = opt.split(":")
+                self.options[key] = value
+
             local = False
             modelFile = request.Model
 
@@ -447,6 +459,9 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         # create a dictionary of parameters by using the keys from EnableParameters and the values from defaults
         kwargs = {key: options.get(key) for key in keys if key in options}
 
+        # populate kwargs from self.options.
+        kwargs.update(self.options)
+
         # Set seed
         if request.seed > 0:
             kwargs["generator"] = torch.Generator(device=self.device).manual_seed(

backend/python/diffusers/requirements.txt

@@ -1,5 +1,5 @@
 setuptools
-grpcio==1.70.0
+grpcio==1.71.0
 pillow
 protobuf
 certifi

backend/python/exllama2/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 certifi
 wheel

backend/python/faster-whisper/requirements.txt

@@ -1,3 +1,3 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 grpcio-tools

backend/python/kokoro/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 phonemizer
 scipy

backend/python/rerankers/requirements.txt

@@ -1,3 +1,3 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 certifi

backend/python/transformers/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 certifi
 setuptools

backend/python/vllm/backend.py

@@ -109,6 +109,17 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
             engine_args.swap_space = request.SwapSpace
         if request.MaxModelLen != 0:
             engine_args.max_model_len = request.MaxModelLen
+        if request.DisableLogStatus:
+            engine_args.disable_log_status = request.DisableLogStatus
+        if request.DType != "":
+            engine_args.dtype = request.DType
+        if request.LimitImagePerPrompt != 0 or request.LimitVideoPerPrompt != 0 or request.LimitAudioPerPrompt != 0:
+            # limit-mm-per-prompt defaults to 1 per modality, based on vLLM docs
+            engine_args.limit_mm_per_prompt = {
+                "image": max(request.LimitImagePerPrompt, 1),
+                "video": max(request.LimitVideoPerPrompt, 1),
+                "audio": max(request.LimitAudioPerPrompt, 1)
+            }
 
         try:
             self.llm = AsyncLLMEngine.from_engine_args(engine_args)
@@ -269,7 +280,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
     def load_image(self, image_path: str):
         """
         Load an image from the given file path or base64 encoded data.
-        
+
         Args:
             image_path (str): The path to the image file or base64 encoded data.
 
@@ -288,7 +299,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
     def load_video(self, video_path: str):
         """
         Load a video from the given file path.
-        
+
         Args:
             video_path (str): The path to the image file.
 
@@ -335,4 +346,4 @@ if __name__ == "__main__":
     )
     args = parser.parse_args()
 
-    asyncio.run(serve(args.addr))
+    asyncio.run(serve(args.addr))

backend/python/vllm/requirements.txt

@@ -1,4 +1,4 @@
-grpcio==1.70.0
+grpcio==1.71.0
 protobuf
 certifi
 setuptools