diff --git a/.env b/.env
index ec53a8f4..c067f7b8 100644
--- a/.env
+++ b/.env
@@ -61,4 +61,9 @@ MODELS_PATH=/models
 # LD_PRELOAD=
 
 ### Huggingface cache for models
-# HUGGINGFACE_HUB_CACHE=/usr/local/huggingface
\ No newline at end of file
+# HUGGINGFACE_HUB_CACHE=/usr/local/huggingface
+
+### Python backends GRPC max workers
+### Default number of workers for GRPC Python backends.
+### This actually controls wether a backend can process multiple requests or not.
+# PYTHON_GRPC_MAX_WORKERS=1
\ No newline at end of file
diff --git a/extra/grpc/autogptq/autogptq.py b/extra/grpc/autogptq/autogptq.py
index 6a5f9c7c..7f0f609f 100755
--- a/extra/grpc/autogptq/autogptq.py
+++ b/extra/grpc/autogptq/autogptq.py
@@ -15,6 +15,9 @@ from transformers import TextGenerationPipeline
 
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 
+# If MAX_WORKERS are specified in the environment use it, otherwise default to 1
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
     def Health(self, request, context):
@@ -77,7 +80,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=1))
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()
diff --git a/extra/grpc/bark/ttsbark.py b/extra/grpc/bark/ttsbark.py
index a14c632d..313dc3a4 100644
--- a/extra/grpc/bark/ttsbark.py
+++ b/extra/grpc/bark/ttsbark.py
@@ -15,6 +15,9 @@ from scipy.io.wavfile import write as write_wav
 
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 
+# If MAX_WORKERS are specified in the environment use it, otherwise default to 1
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
     def Health(self, request, context):
@@ -51,7 +54,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         return backend_pb2.Result(success=True)
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=1))
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()
diff --git a/extra/grpc/diffusers/backend_diffusers.py b/extra/grpc/diffusers/backend_diffusers.py
index c23fc22a..693db1fa 100755
--- a/extra/grpc/diffusers/backend_diffusers.py
+++ b/extra/grpc/diffusers/backend_diffusers.py
@@ -26,6 +26,9 @@ _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 COMPEL=os.environ.get("COMPEL", "1") == "1"
 CLIPSKIP=os.environ.get("CLIPSKIP", "1") == "1"
 
+# If MAX_WORKERS are specified in the environment use it, otherwise default to 1
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
 # https://github.com/CompVis/stable-diffusion/issues/239#issuecomment-1627615287
 def sc(self, clip_input, images) : return images, [False for i in images]
 # edit the StableDiffusionSafetyChecker class so that, when called, it just returns the images and an array of True values
@@ -346,7 +349,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         return backend_pb2.Result(message="Model loaded successfully", success=True)
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=1))
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()
diff --git a/extra/grpc/exllama/exllama.py b/extra/grpc/exllama/exllama.py
index c8eddf4e..25785aae 100755
--- a/extra/grpc/exllama/exllama.py
+++ b/extra/grpc/exllama/exllama.py
@@ -19,6 +19,9 @@ from exllama.tokenizer import ExLlamaTokenizer
 
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 
+# If MAX_WORKERS are specified in the environment use it, otherwise default to 1
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
     def generate(self,prompt, max_new_tokens):
@@ -110,7 +113,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=1))
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()
diff --git a/extra/grpc/huggingface/huggingface.py b/extra/grpc/huggingface/huggingface.py
index 680c2739..8a61b3fa 100755
--- a/extra/grpc/huggingface/huggingface.py
+++ b/extra/grpc/huggingface/huggingface.py
@@ -12,6 +12,9 @@ from sentence_transformers import SentenceTransformer
 
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 
+# If MAX_WORKERS are specified in the environment use it, otherwise default to 1
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
     def Health(self, request, context):
@@ -34,7 +37,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
 
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=1))
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()
diff --git a/extra/grpc/vall-e-x/ttsvalle.py b/extra/grpc/vall-e-x/ttsvalle.py
index 714a3415..be7f3cab 100644
--- a/extra/grpc/vall-e-x/ttsvalle.py
+++ b/extra/grpc/vall-e-x/ttsvalle.py
@@ -16,6 +16,9 @@ from utils.prompt_making import make_prompt
 
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 
+# If MAX_WORKERS are specified in the environment use it, otherwise default to 1
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
     def Health(self, request, context):
@@ -65,7 +68,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         return backend_pb2.Result(success=True)
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=1))
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()
diff --git a/extra/grpc/vllm/backend_vllm.py b/extra/grpc/vllm/backend_vllm.py
index a35cbc74..43231713 100644
--- a/extra/grpc/vllm/backend_vllm.py
+++ b/extra/grpc/vllm/backend_vllm.py
@@ -14,6 +14,9 @@ from vllm import LLM, SamplingParams
 
 _ONE_DAY_IN_SECONDS = 60 * 60 * 24
 
+# If MAX_WORKERS are specified in the environment use it, otherwise default to 1
+MAX_WORKERS = int(os.environ.get('PYTHON_GRPC_MAX_WORKERS', '1'))
+
 # Implement the BackendServicer class with the service methods
 class BackendServicer(backend_pb2_grpc.BackendServicer):
     def generate(self,prompt, max_new_tokens):
@@ -70,7 +73,7 @@ class BackendServicer(backend_pb2_grpc.BackendServicer):
         return self.Predict(request, context)
 
 def serve(address):
-    server = grpc.server(futures.ThreadPoolExecutor(max_workers=1))
+    server = grpc.server(futures.ThreadPoolExecutor(max_workers=MAX_WORKERS))
     backend_pb2_grpc.add_BackendServicer_to_server(BackendServicer(), server)
     server.add_insecure_port(address)
     server.start()