From ea0788d79b2be660834853ed318e86c8eaab7c9e Mon Sep 17 00:00:00 2001
From: Peilin Yu <peilin_yu@brown.edu>
Date: Wed, 19 Jul 2023 13:20:43 -0400
Subject: [PATCH 1/2] Adding vLLM integration

---
 alfred/client/client.py |  5 +++-
 alfred/fm/vllm.py       | 57 +++++++++++++++++++++++++++++++++++++++++
 alfred/run_server.py    |  5 +++-
 3 files changed, 65 insertions(+), 2 deletions(-)
 create mode 100644 alfred/fm/vllm.py

diff --git a/alfred/client/client.py b/alfred/client/client.py
index d3acf14..e7b173b 100644
--- a/alfred/client/client.py
+++ b/alfred/client/client.py
@@ -64,7 +64,7 @@ def __init__(
             assert self.model_type in [
                 "huggingface", "huggingfacevlm",
                 "onnx", "tensorrt",
-                "flexgen",
+                "flexgen", "vllm",
                 "openai", "anthropic",
                 "cohere", "ai21",
                 "torch", "dummy"
@@ -169,6 +169,9 @@ def __init__(
             elif self.model_type == "flexgen":
                 from alfred.fm.flexgen import FlexGenModel
                 self.model = FlexGenModel(self.model, **kwargs)
+            elif self.model_type == "vllm":
+                from alfred.fm.vllm import vLLMModel
+                self.model = vLLMModel(self.model, **kwargs)
             elif self.model_type == "tensorrt":
                 # self.model = TensorRTModel(self.model, **kwargs)
                 raise NotImplementedError
diff --git a/alfred/fm/vllm.py b/alfred/fm/vllm.py
new file mode 100644
index 0000000..df399fc
--- /dev/null
+++ b/alfred/fm/vllm.py
@@ -0,0 +1,57 @@
+import logging
+from typing import List, Any
+
+from alfred.fm.model import LocalAccessFoundationModel
+from .response import CompletionResponse
+
+logger = logging.getLogger(__name__)
+
+
+import torch
+try:
+    from vllm import LLM, SamplingParams
+except ImportError:
+    raise ImportError("Please install VLLM with `pip install vllm`")
+
+
+
+class vLLMModel(LocalAccessFoundationModel):
+    """
+    vLLMModel wraps a vLLM model. vLLM is a fast and easy-to-use library for LLM inference.
+
+    source: https://github.com/vllm-project/vllm
+    """
+
+    def __init__(self, model: str, model_string: str, local_dir: str = None,
+                 **kwargs: Any):
+        """
+        Initialize a VLLM with MultiGPU.
+
+        :param model: (optional) The path to the model.
+        :type model: str
+        """
+        self.model_string = model
+        super().__init__(model_string)
+        self.gpu_count = torch.cuda.device_count()
+        self.model = LLM(local_dir if local_dir is not None else model, tensor_parallel_size=self.gpu_count)
+    def _generate_batch(
+            self,
+            batch_instance: List[str],
+            **kwargs: Any,
+    ) -> List[CompletionResponse]:
+        """
+        Generate completions for a batch of queries.
+
+        :param batch_instance: A list of queries.
+        :type batch_instance: List[str]
+        :param kwargs: Additional keyword arguments.
+        :return: A list of `CompletionResponse` objects with the same prediction content as the input.
+        :rtype: List[CompletionResponse]
+        """
+
+        temperature = kwargs.get("temperature", 0)
+        max_new_tokens = kwargs.get("max_new_tokens", 16)
+
+        sampling_params = SamplingParams(temperature=temperature, max_tokens=max_new_tokens, top_k=1)
+
+        return [CompletionResponse(prediction=output.outputs[0].text) for output in self.model.generate(batch_instance, sampling_params)]
diff --git a/alfred/run_server.py b/alfred/run_server.py
index 028c431..62c470d 100644
--- a/alfred/run_server.py
+++ b/alfred/run_server.py
@@ -45,7 +45,7 @@ def __init__(
         self.model_type = model_type.lower()
         assert self.model_type in [
             "huggingface", "huggingfacevlm", "onnx", "tensorrt", "openai", "anthropic",
-            "flexgen",
+            "flexgen", "vllm",
             "cohere", "ai21", "torch", "dummy"
         ], f"Invalid model type: {self.model_type}"
         if self.model_type == "huggingface":
@@ -77,6 +77,9 @@ def __init__(
         elif self.model_type == "flexgen":
             from alfred.fm.flexgen import FlexGenModel
             self.model = FlexGenModel(self.model, **kwargs)
+        elif self.model_type == "vllm":
+            from alfred.fm.vllm import vLLMModel
+            self.model = vLLMModel(self.model, **kwargs)
         elif self.model_type == "tensorrt":
             # self.model = TensorRTModel(self.model, **kwargs)
             raise NotImplementedError

From 53431db325fe45ff22e134fcb835beb37e798b0d Mon Sep 17 00:00:00 2001
From: Peilin Yu <peilin_yu@brown.edu>
Date: Wed, 19 Jul 2023 13:21:12 -0400
Subject: [PATCH 2/2] Adding vLLM integration Accompany Doc

---
 docs/README.md                                |  1 +
 docs/alfred/client/cache/cache.md             |  4 +---
 docs/alfred/client/cache/dummy.md             |  4 +---
 docs/alfred/client/cache/sqlite.md            |  4 +---
 docs/alfred/client/client.md                  | 20 +++++++++----------
 docs/alfred/client/ssh/sshtunnel.md           |  4 +---
 docs/alfred/client/ssh/utils.md               |  4 +---
 docs/alfred/data/arrow.md                     |  4 +---
 docs/alfred/data/dataset.md                   |  4 +---
 docs/alfred/data/wrench.md                    |  4 +---
 docs/alfred/fm/ai21.md                        |  4 +---
 docs/alfred/fm/anthropic.md                   |  4 +---
 docs/alfred/fm/cohere.md                      |  4 +---
 docs/alfred/fm/dummy.md                       |  4 +---
 docs/alfred/fm/huggingface.md                 |  4 +---
 docs/alfred/fm/huggingfacevlm.md              |  4 +---
 docs/alfred/fm/index.md                       |  3 ++-
 docs/alfred/fm/model.md                       |  4 +---
 docs/alfred/fm/onnx.md                        |  4 +---
 docs/alfred/fm/openai.md                      |  4 +---
 docs/alfred/fm/query/completion_query.md      |  4 +---
 docs/alfred/fm/query/query.md                 |  4 +---
 docs/alfred/fm/query/ranked_query.md          |  4 +---
 docs/alfred/fm/remote/grpc.md                 |  4 +---
 docs/alfred/fm/remote/protos/query_pb2.md     |  1 -
 .../alfred/fm/remote/protos/query_pb2_grpc.md |  4 +---
 docs/alfred/fm/remote/utils.md                |  4 +---
 .../alfred/fm/response/completion_response.md |  4 +---
 docs/alfred/fm/response/ranked_response.md    |  4 +---
 docs/alfred/fm/response/response.md           |  4 +---
 docs/alfred/fm/utils.md                       |  4 +---
 docs/alfred/labeling/flyingsquid.md           |  4 +---
 docs/alfred/labeling/labelmodel.md            |  4 +---
 docs/alfred/labeling/majority_vote.md         |  4 +---
 docs/alfred/labeling/naive_bayes.md           |  4 +---
 docs/alfred/labeling/nplm.md                  |  4 +---
 docs/alfred/run_client_api.md                 |  4 +---
 docs/alfred/run_server.md                     |  6 ++----
 docs/alfred/template/image_template.md        |  4 +---
 docs/alfred/template/string_template.md       |  4 +---
 docs/alfred/template/template.md              |  4 +---
 docs/alfred/voter/voter.md                    |  4 +---
 42 files changed, 51 insertions(+), 128 deletions(-)

diff --git a/docs/README.md b/docs/README.md
index 26bf3a8..63ec788 100644
--- a/docs/README.md
+++ b/docs/README.md
@@ -44,6 +44,7 @@ A full list of `Alfred` project modules.
             - [RankedResponse](alfred/fm/response/ranked_response.md#rankedresponse)
             - [Response](alfred/fm/response/response.md#response)
         - [Utils](alfred/fm/utils.md#utils)
+        - [Vllm](alfred/fm/vllm.md#vllm)
     - [Labeling](alfred/labeling/index.md#labeling)
         - [FlyingSquid](alfred/labeling/flyingsquid.md#flyingsquid)
         - [LabelModel](alfred/labeling/labelmodel.md#labelmodel)
diff --git a/docs/alfred/client/cache/cache.md b/docs/alfred/client/cache/cache.md
index b3d0296..932dfe2 100644
--- a/docs/alfred/client/cache/cache.md
+++ b/docs/alfred/client/cache/cache.md
@@ -372,6 +372,4 @@ Type: *str*
 ```python
 def to_metadata_string(**kwargs: Any) -> str:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/client/cache/dummy.md b/docs/alfred/client/cache/dummy.md
index 19892bf..ecdca0f 100644
--- a/docs/alfred/client/cache/dummy.md
+++ b/docs/alfred/client/cache/dummy.md
@@ -195,6 +195,4 @@ Write a prompt-response pair to the cache
 ```python
 def write(self, prompt: str, response: str, metadata: Optional[str] = None):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/client/cache/sqlite.md b/docs/alfred/client/cache/sqlite.md
index a8422ae..bb5a834 100644
--- a/docs/alfred/client/cache/sqlite.md
+++ b/docs/alfred/client/cache/sqlite.md
@@ -328,6 +328,4 @@ def write_batch(
     self, prompts: List[str], responses: List[str], metadata: Optional[str] = None
 ):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/client/client.md b/docs/alfred/client/client.md
index ad48cd8..c628df6 100644
--- a/docs/alfred/client/client.md
+++ b/docs/alfred/client/client.md
@@ -47,7 +47,7 @@ class Client:
 
 ### Client().__call__
 
-[Show source in client.py:267](../../../alfred/client/client.py#L267)
+[Show source in client.py:270](../../../alfred/client/client.py#L270)
 
 __call__() function to run the model on the queries.
 Equivalent to run() function.
@@ -75,7 +75,7 @@ def __call__(
 
 ### Client().calibrate
 
-[Show source in client.py:282](../../../alfred/client/client.py#L282)
+[Show source in client.py:285](../../../alfred/client/client.py#L285)
 
 calibrate are used to calibrate foundation models contextually given the template.
 A voter class may be passed to calibrate the model with a specific voter.
@@ -120,7 +120,7 @@ def calibrate(
 
 ### Client().chat
 
-[Show source in client.py:384](../../../alfred/client/client.py#L384)
+[Show source in client.py:387](../../../alfred/client/client.py#L387)
 
 Chat with the model APIs.
 Currently, Alfred supports Chat APIs from Anthropic and OpenAI
@@ -139,7 +139,7 @@ def chat(self, log_save_path: Optional[str] = None, **kwargs: Any):
 
 ### Client().encode
 
-[Show source in client.py:358](../../../alfred/client/client.py#L358)
+[Show source in client.py:361](../../../alfred/client/client.py#L361)
 
 embed() function to embed the queries.
 
@@ -162,7 +162,7 @@ def encode(
 
 ### Client().generate
 
-[Show source in client.py:226](../../../alfred/client/client.py#L226)
+[Show source in client.py:229](../../../alfred/client/client.py#L229)
 
 Wrapper function to generate the response(s) from the model. (For completion)
 
@@ -191,7 +191,7 @@ def generate(
 
 ### Client().remote_run
 
-[Show source in client.py:204](../../../alfred/client/client.py#L204)
+[Show source in client.py:207](../../../alfred/client/client.py#L207)
 
 Wrapper function for running the model on the queries thru a gRPC Server.
 
@@ -218,7 +218,7 @@ def remote_run(
 
 ### Client().run
 
-[Show source in client.py:184](../../../alfred/client/client.py#L184)
+[Show source in client.py:187](../../../alfred/client/client.py#L187)
 
 Run the model on the queries.
 
@@ -245,7 +245,7 @@ def run(
 
 ### Client().score
 
-[Show source in client.py:243](../../../alfred/client/client.py#L243)
+[Show source in client.py:246](../../../alfred/client/client.py#L246)
 
 Wrapper function to score the response(s) from the model. (For ranking)
 
@@ -275,6 +275,4 @@ def score(
     self, query: Union[RankedQuery, Dict, List[RankedQuery], List[str]], **kwargs: Any
 ) -> Union[Response, List[Response]]:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/client/ssh/sshtunnel.md b/docs/alfred/client/ssh/sshtunnel.md
index 36c88ab..ca7dec9 100644
--- a/docs/alfred/client/ssh/sshtunnel.md
+++ b/docs/alfred/client/ssh/sshtunnel.md
@@ -79,6 +79,4 @@ Stop the tunnel
 ```python
 def stop(self):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/client/ssh/utils.md b/docs/alfred/client/ssh/utils.md
index c918781..f17444f 100644
--- a/docs/alfred/client/ssh/utils.md
+++ b/docs/alfred/client/ssh/utils.md
@@ -94,6 +94,4 @@ Finds the next available port if given port is not available
 ```python
 def port_finder(port: Union[str, int], host: str = "") -> int:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/data/arrow.md b/docs/alfred/data/arrow.md
index af302df..0dec50f 100644
--- a/docs/alfred/data/arrow.md
+++ b/docs/alfred/data/arrow.md
@@ -614,6 +614,4 @@ returns the version of the dataset
 ```python
 def version(self) -> str:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/data/dataset.md b/docs/alfred/data/dataset.md
index 3ddc8d5..8fc9cca 100644
--- a/docs/alfred/data/dataset.md
+++ b/docs/alfred/data/dataset.md
@@ -194,6 +194,4 @@ returns the version of the dataset
 @property
 def version(self) -> str:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/data/wrench.md b/docs/alfred/data/wrench.md
index 08e3213..5600925 100644
--- a/docs/alfred/data/wrench.md
+++ b/docs/alfred/data/wrench.md
@@ -64,6 +64,4 @@ returns the string representation of the dataset
 ```python
 def __repr__(self):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/ai21.md b/docs/alfred/fm/ai21.md
index 0d0920a..fa32f4e 100644
--- a/docs/alfred/fm/ai21.md
+++ b/docs/alfred/fm/ai21.md
@@ -24,6 +24,4 @@ This class provides a wrapper for the OpenAI API for generating completions.
 class AI21Model(APIAccessFoundationModel):
     def __init__(self, model_string: str = "j1-large", api_key: Optional[str] = None):
         ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/anthropic.md b/docs/alfred/fm/anthropic.md
index b64b19f..6d047ab 100644
--- a/docs/alfred/fm/anthropic.md
+++ b/docs/alfred/fm/anthropic.md
@@ -38,6 +38,4 @@ Launch an interactive chat session with the Anthropic API.
 ```python
 def chat(self, **kwargs: Any):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/cohere.md b/docs/alfred/fm/cohere.md
index 8e73833..03b6b87 100644
--- a/docs/alfred/fm/cohere.md
+++ b/docs/alfred/fm/cohere.md
@@ -24,6 +24,4 @@ This class provides a wrapper for the OpenAI API for generating completions.
 class CohereModel(APIAccessFoundationModel):
     def __init__(self, model_string: str = "xlarge", api_key: Optional[str] = None):
         ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/dummy.md b/docs/alfred/fm/dummy.md
index 454f0ba..b40d198 100644
--- a/docs/alfred/fm/dummy.md
+++ b/docs/alfred/fm/dummy.md
@@ -29,6 +29,4 @@ class DummyModel(LocalAccessFoundationModel):
 
 #### See also
 
-- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
-
-
+- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
\ No newline at end of file
diff --git a/docs/alfred/fm/huggingface.md b/docs/alfred/fm/huggingface.md
index b5d6d16..104c723 100644
--- a/docs/alfred/fm/huggingface.md
+++ b/docs/alfred/fm/huggingface.md
@@ -45,6 +45,4 @@ class HuggingFaceModel(LocalAccessFoundationModel):
 
 #### See also
 
-- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
-
-
+- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
\ No newline at end of file
diff --git a/docs/alfred/fm/huggingfacevlm.md b/docs/alfred/fm/huggingfacevlm.md
index 77a1677..51473ce 100644
--- a/docs/alfred/fm/huggingfacevlm.md
+++ b/docs/alfred/fm/huggingfacevlm.md
@@ -33,6 +33,4 @@ class HuggingFaceCLIPModel(LocalAccessFoundationModel):
 
 #### See also
 
-- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
-
-
+- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
\ No newline at end of file
diff --git a/docs/alfred/fm/index.md b/docs/alfred/fm/index.md
index 4add719..e1c7cd6 100644
--- a/docs/alfred/fm/index.md
+++ b/docs/alfred/fm/index.md
@@ -24,4 +24,5 @@ Fm
 - [Query](query/index.md)
 - [Remote](remote/index.md)
 - [Response](response/index.md)
-- [Utils](./utils.md)
\ No newline at end of file
+- [Utils](./utils.md)
+- [Vllm](./vllm.md)
\ No newline at end of file
diff --git a/docs/alfred/fm/model.md b/docs/alfred/fm/model.md
index 1778933..710a22c 100644
--- a/docs/alfred/fm/model.md
+++ b/docs/alfred/fm/model.md
@@ -286,6 +286,4 @@ class LocalAccessFoundationModel(FoundationModel):
 
 #### See also
 
-- [FoundationModel](#foundationmodel)
-
-
+- [FoundationModel](#foundationmodel)
\ No newline at end of file
diff --git a/docs/alfred/fm/onnx.md b/docs/alfred/fm/onnx.md
index f21acab..449923b 100644
--- a/docs/alfred/fm/onnx.md
+++ b/docs/alfred/fm/onnx.md
@@ -30,6 +30,4 @@ class ONNXModel(LocalAccessFoundationModel):
 
 #### See also
 
-- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
-
-
+- [LocalAccessFoundationModel](./model.md#localaccessfoundationmodel)
\ No newline at end of file
diff --git a/docs/alfred/fm/openai.md b/docs/alfred/fm/openai.md
index 3d0e6c4..a127a2b 100644
--- a/docs/alfred/fm/openai.md
+++ b/docs/alfred/fm/openai.md
@@ -40,6 +40,4 @@ Launch an interactive chat session with the OpenAI API.
 ```python
 def chat(self, **kwargs: Any):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/query/completion_query.md b/docs/alfred/fm/query/completion_query.md
index 8339e34..b0e98c8 100644
--- a/docs/alfred/fm/query/completion_query.md
+++ b/docs/alfred/fm/query/completion_query.md
@@ -143,6 +143,4 @@ returns the raw prompt content
 @property
 def prompt(self):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/query/query.md b/docs/alfred/fm/query/query.md
index 92c0da6..acbdcc4 100644
--- a/docs/alfred/fm/query/query.md
+++ b/docs/alfred/fm/query/query.md
@@ -82,6 +82,4 @@ Type: *str*
 ```python
 def serialize(self) -> str:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/query/ranked_query.md b/docs/alfred/fm/query/ranked_query.md
index c1c68f1..57bd2ec 100644
--- a/docs/alfred/fm/query/ranked_query.md
+++ b/docs/alfred/fm/query/ranked_query.md
@@ -195,6 +195,4 @@ returns the raw prompt content
 @property
 def prompt(self):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/remote/grpc.md b/docs/alfred/fm/remote/grpc.md
index b921839..dfa9880 100644
--- a/docs/alfred/fm/remote/grpc.md
+++ b/docs/alfred/fm/remote/grpc.md
@@ -132,6 +132,4 @@ def restart(self):
 ```python
 def serve(self, credentials: Optional[grpc.ServerCredentials] = None):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/remote/protos/query_pb2.md b/docs/alfred/fm/remote/protos/query_pb2.md
index 841a283..014445b 100644
--- a/docs/alfred/fm/remote/protos/query_pb2.md
+++ b/docs/alfred/fm/remote/protos/query_pb2.md
@@ -8,5 +8,4 @@
 Query Pb2
 
 > Auto-generated documentation for [alfred.fm.remote.protos.query_pb2](../../../../../alfred/fm/remote/protos/query_pb2.py) module.
-
 - [Query Pb2](#query-pb2)
diff --git a/docs/alfred/fm/remote/protos/query_pb2_grpc.md b/docs/alfred/fm/remote/protos/query_pb2_grpc.md
index b1fb984..781e262 100644
--- a/docs/alfred/fm/remote/protos/query_pb2_grpc.md
+++ b/docs/alfred/fm/remote/protos/query_pb2_grpc.md
@@ -146,6 +146,4 @@ class QueryServiceStub(object):
 ```python
 def add_QueryServiceServicer_to_server(servicer, server):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/remote/utils.md b/docs/alfred/fm/remote/utils.md
index ed39c98..f97bf67 100644
--- a/docs/alfred/fm/remote/utils.md
+++ b/docs/alfred/fm/remote/utils.md
@@ -76,6 +76,4 @@ def port_finder(port: int) -> int:
 ```python
 def tensor_to_bytes(tensor):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/response/completion_response.md b/docs/alfred/fm/response/completion_response.md
index cdfac42..3277ba5 100644
--- a/docs/alfred/fm/response/completion_response.md
+++ b/docs/alfred/fm/response/completion_response.md
@@ -120,6 +120,4 @@ Type: *float*
 @property
 def score(self) -> Dict:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/response/ranked_response.md b/docs/alfred/fm/response/ranked_response.md
index 4c9e97f..cf1e961 100644
--- a/docs/alfred/fm/response/ranked_response.md
+++ b/docs/alfred/fm/response/ranked_response.md
@@ -137,6 +137,4 @@ Type: *dict*
 @property
 def scores(self) -> Dict:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/response/response.md b/docs/alfred/fm/response/response.md
index 55d556e..389184c 100644
--- a/docs/alfred/fm/response/response.md
+++ b/docs/alfred/fm/response/response.md
@@ -100,6 +100,4 @@ Type: *str*
 ```python
 def serialize(self) -> str:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/fm/utils.md b/docs/alfred/fm/utils.md
index ff15a19..3825d7c 100644
--- a/docs/alfred/fm/utils.md
+++ b/docs/alfred/fm/utils.md
@@ -361,6 +361,4 @@ Type: *List[int]*
 ```python
 def tokenize(inst, tokenizer, max_length=512):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/labeling/flyingsquid.md b/docs/alfred/labeling/flyingsquid.md
index 73dae10..eec789f 100644
--- a/docs/alfred/labeling/flyingsquid.md
+++ b/docs/alfred/labeling/flyingsquid.md
@@ -34,6 +34,4 @@ class FlyingSquid(LabelModel):
 ```python
 def label(self, votes: np.ndarray) -> np.ndarray:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/labeling/labelmodel.md b/docs/alfred/labeling/labelmodel.md
index 291f46e..bea20d8 100644
--- a/docs/alfred/labeling/labelmodel.md
+++ b/docs/alfred/labeling/labelmodel.md
@@ -49,6 +49,4 @@ def __call__(self, votes):
 @abc.abstractmethod
 def label(self, votes):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/labeling/majority_vote.md b/docs/alfred/labeling/majority_vote.md
index 56dbf7e..94294e2 100644
--- a/docs/alfred/labeling/majority_vote.md
+++ b/docs/alfred/labeling/majority_vote.md
@@ -36,6 +36,4 @@ returns the majority vote for each response row
 ```python
 def label(self, votes: np.ndarray) -> np.ndarray:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/labeling/naive_bayes.md b/docs/alfred/labeling/naive_bayes.md
index 9662eab..5b24189 100644
--- a/docs/alfred/labeling/naive_bayes.md
+++ b/docs/alfred/labeling/naive_bayes.md
@@ -47,6 +47,4 @@ Type: *np.ndarray*
 ```python
 def label(self, votes: np.ndarray) -> np.ndarray:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/labeling/nplm.md b/docs/alfred/labeling/nplm.md
index 6d7be6e..132833a 100644
--- a/docs/alfred/labeling/nplm.md
+++ b/docs/alfred/labeling/nplm.md
@@ -52,6 +52,4 @@ Type: *np.ndarray*
 ```python
 def label(self, votes: np.ndarray) -> np.ndarray:
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/run_client_api.md b/docs/alfred/run_client_api.md
index 82faf32..d1ea282 100644
--- a/docs/alfred/run_client_api.md
+++ b/docs/alfred/run_client_api.md
@@ -366,6 +366,4 @@ async def set_alfred_server_webhook_port(request: Request):
 @alfred_app.get("/status")
 async def status():
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/run_server.md b/docs/alfred/run_server.md
index b73e160..9b75516 100644
--- a/docs/alfred/run_server.md
+++ b/docs/alfred/run_server.md
@@ -29,7 +29,7 @@ class ModelServer:
 
 ## start_server
 
-[Show source in run_server.py:96](../../alfred/run_server.py#L96)
+[Show source in run_server.py:99](../../alfred/run_server.py#L99)
 
 Wrapper function to start gRPC Server.
 
@@ -43,6 +43,4 @@ Wrapper function to start gRPC Server.
 ```python
 def start_server(args: argparse.Namespace):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/template/image_template.md b/docs/alfred/template/image_template.md
index 7f57260..b3f7c99 100644
--- a/docs/alfred/template/image_template.md
+++ b/docs/alfred/template/image_template.md
@@ -293,6 +293,4 @@ returns the template type
 @property
 def type(self):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/template/string_template.md b/docs/alfred/template/string_template.md
index 8790f42..cde3ca0 100644
--- a/docs/alfred/template/string_template.md
+++ b/docs/alfred/template/string_template.md
@@ -334,6 +334,4 @@ returns the template type
 @property
 def type(self):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/template/template.md b/docs/alfred/template/template.md
index 9f17393..1377eb9 100644
--- a/docs/alfred/template/template.md
+++ b/docs/alfred/template/template.md
@@ -219,6 +219,4 @@ returns the type of the template
 @abc.abstractmethod
 def type(self):
     ...
-```
-
-
+```
\ No newline at end of file
diff --git a/docs/alfred/voter/voter.md b/docs/alfred/voter/voter.md
index b61a0c4..760c81e 100644
--- a/docs/alfred/voter/voter.md
+++ b/docs/alfred/voter/voter.md
@@ -114,6 +114,4 @@ def vote(
 
 #### See also
 
-- [Response](../fm/response/response.md#response)
-
-
+- [Response](../fm/response/response.md#response)
\ No newline at end of file