From 8eec0c8a1a0e6bb13b7ed942691de8b672631ae7 Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Tue, 12 Nov 2024 16:16:12 +0800
Subject: [PATCH 01/18] Embedding TEI Langchain compatible with OpenAI API

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 comps/embeddings/tei/langchain/README.md        | 16 ++++++++++++++++
 comps/embeddings/tei/langchain/embedding_tei.py | 17 +++++++++--------
 2 files changed, 25 insertions(+), 8 deletions(-)

diff --git a/comps/embeddings/tei/langchain/README.md b/comps/embeddings/tei/langchain/README.md
index 05705c4bb..61219af83 100644
--- a/comps/embeddings/tei/langchain/README.md
+++ b/comps/embeddings/tei/langchain/README.md
@@ -113,9 +113,25 @@ curl http://localhost:6000/v1/health_check\
 
 ### 3.2 Consume Embedding Service
 
+Basic API, query with single text.
 ```bash
 curl http://localhost:6000/v1/embeddings\
   -X POST \
   -d '{"text":"Hello, world!"}' \
   -H 'Content-Type: application/json'
 ```
+
+We are also compatible with [OpenAI API](https://platform.openai.com/docs/api-reference/embeddings).
+```bash
+## Input single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
+
+## Input multiple texts with parameters
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":["Hello, world!","How are you?"], "dimensions":100}' \
+  -H 'Content-Type: application/json'
+```
diff --git a/comps/embeddings/tei/langchain/embedding_tei.py b/comps/embeddings/tei/langchain/embedding_tei.py
index 02596f467..3cab975f5 100644
--- a/comps/embeddings/tei/langchain/embedding_tei.py
+++ b/comps/embeddings/tei/langchain/embedding_tei.py
@@ -57,19 +57,19 @@ async def embedding(
         logger.info(input)
     if isinstance(input, TextDoc):
         embed_vector = await aembed_query(input.text, async_client)
-        res = EmbedDoc(text=input.text, embedding=embed_vector)
+        res = EmbedDoc(text=input.text, embedding=embed_vector[0])
     else:
         embed_vector = await aembed_query(input.input, async_client)
         if input.dimensions is not None:
-            embed_vector = embed_vector[: input.dimensions]
+            embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
+
+        # for standard openai embedding format
+        res = EmbeddingResponse(data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))])
 
         if isinstance(input, ChatCompletionRequest):
-            input.embedding = embed_vector
+            input.embedding = res
             # keep
             res = input
-        if isinstance(input, EmbeddingRequest):
-            # for standard openai embedding format
-            res = EmbeddingResponse(data=[EmbeddingResponseData(index=0, embedding=embed_vector)])
 
     statistics_dict["opea_service@embedding_tei_langchain"].append_latency(time.time() - start, None)
     if logflag:
@@ -77,8 +77,9 @@ async def embedding(
     return res
 
 
-async def aembed_query(text: str, async_client: AsyncInferenceClient, model_kwargs=None, task=None) -> List[float]:
-    response = (await aembed_documents([text], async_client, model_kwargs=model_kwargs, task=task))[0]
+async def aembed_query(text: Union[str, List[str]], async_client: AsyncInferenceClient, model_kwargs=None, task=None) -> List[List[float]]:
+    texts=[text] if isinstance(text, str) else text
+    response = (await aembed_documents(texts, async_client, model_kwargs=model_kwargs, task=task))
     return response
 
 

From cddd0bdbc21b36d8e863ee5496d971090a15369d Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Tue, 12 Nov 2024 08:56:58 +0000
Subject: [PATCH 02/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 comps/embeddings/tei/langchain/README.md        |  2 ++
 comps/embeddings/tei/langchain/embedding_tei.py | 12 ++++++++----
 2 files changed, 10 insertions(+), 4 deletions(-)

diff --git a/comps/embeddings/tei/langchain/README.md b/comps/embeddings/tei/langchain/README.md
index 61219af83..95b78511b 100644
--- a/comps/embeddings/tei/langchain/README.md
+++ b/comps/embeddings/tei/langchain/README.md
@@ -114,6 +114,7 @@ curl http://localhost:6000/v1/health_check\
 ### 3.2 Consume Embedding Service
 
 Basic API, query with single text.
+
 ```bash
 curl http://localhost:6000/v1/embeddings\
   -X POST \
@@ -122,6 +123,7 @@ curl http://localhost:6000/v1/embeddings\
 ```
 
 We are also compatible with [OpenAI API](https://platform.openai.com/docs/api-reference/embeddings).
+
 ```bash
 ## Input single text
 curl http://localhost:6000/v1/embeddings\
diff --git a/comps/embeddings/tei/langchain/embedding_tei.py b/comps/embeddings/tei/langchain/embedding_tei.py
index 3cab975f5..f4f98b8b8 100644
--- a/comps/embeddings/tei/langchain/embedding_tei.py
+++ b/comps/embeddings/tei/langchain/embedding_tei.py
@@ -64,7 +64,9 @@ async def embedding(
             embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
 
         # for standard openai embedding format
-        res = EmbeddingResponse(data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))])
+        res = EmbeddingResponse(
+            data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))]
+        )
 
         if isinstance(input, ChatCompletionRequest):
             input.embedding = res
@@ -77,9 +79,11 @@ async def embedding(
     return res
 
 
-async def aembed_query(text: Union[str, List[str]], async_client: AsyncInferenceClient, model_kwargs=None, task=None) -> List[List[float]]:
-    texts=[text] if isinstance(text, str) else text
-    response = (await aembed_documents(texts, async_client, model_kwargs=model_kwargs, task=task))
+async def aembed_query(
+    text: Union[str, List[str]], async_client: AsyncInferenceClient, model_kwargs=None, task=None
+) -> List[List[float]]:
+    texts = [text] if isinstance(text, str) else text
+    response = await aembed_documents(texts, async_client, model_kwargs=model_kwargs, task=task)
     return response
 
 

From 763e927f9d555d7edda3c98a10798b5a0e7e2a30 Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 09:15:33 +0800
Subject: [PATCH 03/18] TextDoc support list

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 comps/cores/proto/docarray.py                   | 6 +++---
 comps/embeddings/tei/langchain/README.md        | 9 ++++++++-
 comps/embeddings/tei/langchain/embedding_tei.py | 3 ++-
 3 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/comps/cores/proto/docarray.py b/comps/cores/proto/docarray.py
index 71b6f15ec..12465dd38 100644
--- a/comps/cores/proto/docarray.py
+++ b/comps/cores/proto/docarray.py
@@ -17,7 +17,7 @@ class TopologyInfo:
 
 
 class TextDoc(BaseDoc, TopologyInfo):
-    text: str = None
+    text: Union[str, List[str]] = None
 
 
 class FactualityDoc(BaseDoc):
@@ -83,8 +83,8 @@ class DocPath(BaseDoc):
 
 
 class EmbedDoc(BaseDoc):
-    text: str
-    embedding: conlist(float, min_length=0)
+    text: Union[str, List[str]]
+    embedding: Union[conlist(float, min_length=0),List[conlist(float, min_length=0)]]
     search_type: str = "similarity"
     k: int = 4
     distance_threshold: Optional[float] = None
diff --git a/comps/embeddings/tei/langchain/README.md b/comps/embeddings/tei/langchain/README.md
index 95b78511b..96163c915 100644
--- a/comps/embeddings/tei/langchain/README.md
+++ b/comps/embeddings/tei/langchain/README.md
@@ -113,13 +113,20 @@ curl http://localhost:6000/v1/health_check\
 
 ### 3.2 Consume Embedding Service
 
-Basic API, query with single text.
+Use our basic API.
 
 ```bash
+## query with single text
 curl http://localhost:6000/v1/embeddings\
   -X POST \
   -d '{"text":"Hello, world!"}' \
   -H 'Content-Type: application/json'
+
+## query with multiple texts
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":["Hello, world!","How are you?"]}' \
+  -H 'Content-Type: application/json'
 ```
 
 We are also compatible with [OpenAI API](https://platform.openai.com/docs/api-reference/embeddings).
diff --git a/comps/embeddings/tei/langchain/embedding_tei.py b/comps/embeddings/tei/langchain/embedding_tei.py
index f4f98b8b8..20e61196d 100644
--- a/comps/embeddings/tei/langchain/embedding_tei.py
+++ b/comps/embeddings/tei/langchain/embedding_tei.py
@@ -57,7 +57,8 @@ async def embedding(
         logger.info(input)
     if isinstance(input, TextDoc):
         embed_vector = await aembed_query(input.text, async_client)
-        res = EmbedDoc(text=input.text, embedding=embed_vector[0])
+        embedding_res = embed_vector[0] if isinstance(input.text, str) else embed_vector
+        res = EmbedDoc(text=input.text, embedding=embedding_res)
     else:
         embed_vector = await aembed_query(input.input, async_client)
         if input.dimensions is not None:

From 0a1edb0ccbfe4293ffd1f340916c2c094465f152 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 01:55:24 +0000
Subject: [PATCH 04/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 comps/cores/proto/docarray.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/comps/cores/proto/docarray.py b/comps/cores/proto/docarray.py
index 12465dd38..10031e478 100644
--- a/comps/cores/proto/docarray.py
+++ b/comps/cores/proto/docarray.py
@@ -84,7 +84,7 @@ class DocPath(BaseDoc):
 
 class EmbedDoc(BaseDoc):
     text: Union[str, List[str]]
-    embedding: Union[conlist(float, min_length=0),List[conlist(float, min_length=0)]]
+    embedding: Union[conlist(float, min_length=0), List[conlist(float, min_length=0)]]
     search_type: str = "similarity"
     k: int = 4
     distance_threshold: Optional[float] = None

From d5bd70ec3fd2edcdfe3e092cf283be670e6540be Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 09:55:43 +0800
Subject: [PATCH 05/18] support tei llama index openai compatible API

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 comps/embeddings/tei/llama_index/README.md    | 25 +++++++++++++
 .../tei/llama_index/embedding_tei.py          | 37 +++++++++++++++++--
 2 files changed, 58 insertions(+), 4 deletions(-)

diff --git a/comps/embeddings/tei/llama_index/README.md b/comps/embeddings/tei/llama_index/README.md
index e466c47ea..019b55ad3 100644
--- a/comps/embeddings/tei/llama_index/README.md
+++ b/comps/embeddings/tei/llama_index/README.md
@@ -113,9 +113,34 @@ curl http://localhost:6000/v1/health_check\
 
 ### 3.2 Consume Embedding Service
 
+Use our basic API.
+
 ```bash
+## query with single text
 curl http://localhost:6000/v1/embeddings\
   -X POST \
   -d '{"text":"Hello, world!"}' \
   -H 'Content-Type: application/json'
+
+## query with multiple texts
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":["Hello, world!","How are you?"]}' \
+  -H 'Content-Type: application/json'
 ```
+
+We are also compatible with [OpenAI API](https://platform.openai.com/docs/api-reference/embeddings).
+
+```bash
+## Input single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
+
+## Input multiple texts with parameters
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":["Hello, world!","How are you?"], "dimensions":100}' \
+  -H 'Content-Type: application/json'
+```
\ No newline at end of file
diff --git a/comps/embeddings/tei/llama_index/embedding_tei.py b/comps/embeddings/tei/llama_index/embedding_tei.py
index e96b75e75..75238c3a0 100644
--- a/comps/embeddings/tei/llama_index/embedding_tei.py
+++ b/comps/embeddings/tei/llama_index/embedding_tei.py
@@ -2,10 +2,16 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import os
-
+from typing import List, Union
 from llama_index.embeddings.text_embeddings_inference import TextEmbeddingsInference
 
 from comps import CustomLogger, EmbedDoc, ServiceType, TextDoc, opea_microservices, register_microservice
+from comps.cores.proto.api_protocol import (
+    ChatCompletionRequest,
+    EmbeddingRequest,
+    EmbeddingResponse,
+    EmbeddingResponseData,
+)
 
 logger = CustomLogger("embedding_tei_llamaindex")
 logflag = os.getenv("LOGFLAG", False)
@@ -20,15 +26,38 @@
     input_datatype=TextDoc,
     output_datatype=EmbedDoc,
 )
-async def embedding(input: TextDoc) -> EmbedDoc:
+async def embedding(
+    input: Union[TextDoc, EmbeddingRequest, ChatCompletionRequest]
+) -> Union[EmbedDoc, EmbeddingResponse, ChatCompletionRequest]:
     if logflag:
         logger.info(input)
-    embed_vector = await embeddings.aget_query_embedding(input.text)
-    res = EmbedDoc(text=input.text, embedding=embed_vector)
+    if isinstance(input, TextDoc):
+        embed_vector = await get_embeddings(input.text)
+        embedding_res = embed_vector[0] if isinstance(input.text, str) else embed_vector
+        res = EmbedDoc(text=input.text, embedding=embedding_res)
+    else:
+        embed_vector = await get_embeddings(input.input)
+        if input.dimensions is not None:
+            embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
+        
+        # for standard openai embedding format
+        res = EmbeddingResponse(
+            data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))]
+        )
+
+        if isinstance(input, ChatCompletionRequest):
+            input.embedding = res
+            # keep
+            res = input
+    
     if logflag:
         logger.info(res)
     return res
 
+async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
+    texts = [text] if isinstance(text, str) else text
+    embed_vector = await embeddings._aget_text_embeddings(texts)
+    return embed_vector
 
 if __name__ == "__main__":
     tei_embedding_model_name = os.getenv("TEI_EMBEDDING_MODEL_NAME", "BAAI/bge-base-en-v1.5")

From a9173367b659650142f95f62ffdd9e70ebf4bc4d Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 02:35:51 +0000
Subject: [PATCH 06/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 comps/embeddings/tei/llama_index/README.md        | 2 +-
 comps/embeddings/tei/llama_index/embedding_tei.py | 7 +++++--
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/comps/embeddings/tei/llama_index/README.md b/comps/embeddings/tei/llama_index/README.md
index 019b55ad3..dd1f5006c 100644
--- a/comps/embeddings/tei/llama_index/README.md
+++ b/comps/embeddings/tei/llama_index/README.md
@@ -143,4 +143,4 @@ curl http://localhost:6000/v1/embeddings\
   -X POST \
   -d '{"input":["Hello, world!","How are you?"], "dimensions":100}' \
   -H 'Content-Type: application/json'
-```
\ No newline at end of file
+```
diff --git a/comps/embeddings/tei/llama_index/embedding_tei.py b/comps/embeddings/tei/llama_index/embedding_tei.py
index 75238c3a0..a3ff25a70 100644
--- a/comps/embeddings/tei/llama_index/embedding_tei.py
+++ b/comps/embeddings/tei/llama_index/embedding_tei.py
@@ -3,6 +3,7 @@
 
 import os
 from typing import List, Union
+
 from llama_index.embeddings.text_embeddings_inference import TextEmbeddingsInference
 
 from comps import CustomLogger, EmbedDoc, ServiceType, TextDoc, opea_microservices, register_microservice
@@ -39,7 +40,7 @@ async def embedding(
         embed_vector = await get_embeddings(input.input)
         if input.dimensions is not None:
             embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
-        
+
         # for standard openai embedding format
         res = EmbeddingResponse(
             data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))]
@@ -49,16 +50,18 @@ async def embedding(
             input.embedding = res
             # keep
             res = input
-    
+
     if logflag:
         logger.info(res)
     return res
 
+
 async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
     texts = [text] if isinstance(text, str) else text
     embed_vector = await embeddings._aget_text_embeddings(texts)
     return embed_vector
 
+
 if __name__ == "__main__":
     tei_embedding_model_name = os.getenv("TEI_EMBEDDING_MODEL_NAME", "BAAI/bge-base-en-v1.5")
     tei_embedding_endpoint = os.getenv("TEI_EMBEDDING_ENDPOINT", "http://localhost:8090")

From 2fb3d54c19c99d7e8aef6ebf3dcccc85b283c9c3 Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 13:00:16 +0800
Subject: [PATCH 07/18] support mosec langchain openai compatible API

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 comps/embeddings/mosec/langchain/README.md    | 35 ++++++++++++++---
 .../mosec/langchain/embedding_mosec.py        | 38 ++++++++++++++++---
 2 files changed, 63 insertions(+), 10 deletions(-)

diff --git a/comps/embeddings/mosec/langchain/README.md b/comps/embeddings/mosec/langchain/README.md
index 88c888a97..2ea3f32bc 100644
--- a/comps/embeddings/mosec/langchain/README.md
+++ b/comps/embeddings/mosec/langchain/README.md
@@ -25,9 +25,34 @@ docker run -d --name="embedding-langchain-mosec-server" -e http_proxy=$http_prox
 
 ## run client test
 
-```
-curl localhost:6000/v1/embeddings \
-     -X POST \
-     -d '{"text":"Hello, world!"}' \
-     -H 'Content-Type: application/json'
+Use our basic API.
+
+```bash
+## query with single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
+
+## query with multiple texts
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":["Hello, world!","How are you?"]}' \
+  -H 'Content-Type: application/json'
+```
+
+We are also compatible with [OpenAI API](https://platform.openai.com/docs/api-reference/embeddings).
+
+```bash
+## Input single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
+
+## Input multiple texts with parameters
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":["Hello, world!","How are you?"], "dimensions":100}' \
+  -H 'Content-Type: application/json'
 ```
diff --git a/comps/embeddings/mosec/langchain/embedding_mosec.py b/comps/embeddings/mosec/langchain/embedding_mosec.py
index fde9e17af..2f6800277 100644
--- a/comps/embeddings/mosec/langchain/embedding_mosec.py
+++ b/comps/embeddings/mosec/langchain/embedding_mosec.py
@@ -4,7 +4,7 @@
 import asyncio
 import os
 import time
-from typing import List, Optional
+from typing import List, Optional, Union
 
 from langchain_community.embeddings import OpenAIEmbeddings
 
@@ -18,7 +18,12 @@
     register_statistics,
     statistics_dict,
 )
-
+from comps.cores.proto.api_protocol import (
+    ChatCompletionRequest,
+    EmbeddingRequest,
+    EmbeddingResponse,
+    EmbeddingResponseData,
+)
 logger = CustomLogger("embedding_mosec")
 logflag = os.getenv("LOGFLAG", False)
 
@@ -62,17 +67,40 @@ async def get_embedding(e: Optional[List[float]]) -> List[float]:
     output_datatype=EmbedDoc,
 )
 @register_statistics(names=["opea_service@embedding_mosec"])
-async def embedding(input: TextDoc) -> EmbedDoc:
+async def embedding(
+    input: Union[TextDoc, EmbeddingRequest, ChatCompletionRequest]
+) -> Union[EmbedDoc, EmbeddingResponse, ChatCompletionRequest]:
     if logflag:
         logger.info(input)
     start = time.time()
-    embed_vector = await embeddings.aembed_query(input.text)
-    res = EmbedDoc(text=input.text, embedding=embed_vector)
+    if isinstance(input, TextDoc):
+        embed_vector = await get_embeddings(input.text)
+        embedding_res = embed_vector[0] if isinstance(input.text, str) else embed_vector
+        res = EmbedDoc(text=input.text, embedding=embedding_res)
+    else:
+        embed_vector = await get_embeddings(input.input)
+        if input.dimensions is not None:
+            embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
+        
+        # for standard openai embedding format
+        res = EmbeddingResponse(
+            data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))]
+        )
+
+        if isinstance(input, ChatCompletionRequest):
+            input.embedding = res
+            # keep
+            res = input
+
     statistics_dict["opea_service@embedding_mosec"].append_latency(time.time() - start, None)
     if logflag:
         logger.info(res)
     return res
 
+async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
+    texts = [text] if isinstance(text, str) else text
+    embed_vector = await embeddings.aembed_documents(texts)
+    return embed_vector
 
 if __name__ == "__main__":
     MOSEC_EMBEDDING_ENDPOINT = os.environ.get("MOSEC_EMBEDDING_ENDPOINT", "http://127.0.0.1:8080")

From b62b75e16a86162ea3c8fcf5b933ea7ccb833ea0 Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 13:35:41 +0800
Subject: [PATCH 08/18] update UT for embedding tests

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 .../test_embeddings_mosec_langchain.sh        | 22 ++++++++++++++--
 .../test_embeddings_tei_langchain.sh          | 23 ++++++++++++++--
 .../test_embeddings_tei_llama_index.sh        | 26 ++++++++++++++++---
 3 files changed, 64 insertions(+), 7 deletions(-)

diff --git a/tests/embeddings/test_embeddings_mosec_langchain.sh b/tests/embeddings/test_embeddings_mosec_langchain.sh
index 4140e7011..d99d656b0 100644
--- a/tests/embeddings/test_embeddings_mosec_langchain.sh
+++ b/tests/embeddings/test_embeddings_mosec_langchain.sh
@@ -42,11 +42,12 @@ function start_service() {
     sleep 3m
 }
 
-function validate_microservice() {
+function validate_service() {
+    local INPUT_DATA="$1"
     mosec_service_port=5002
     http_proxy="" curl http://${ip_address}:$mosec_service_port/v1/embeddings \
         -X POST \
-        -d '{"text":"What is Deep Learning?"}' \
+        -d "$INPUT_DATA" \
         -H 'Content-Type: application/json'
     if [ $? -eq 0 ]; then
         echo "curl command executed successfully"
@@ -56,6 +57,23 @@ function validate_microservice() {
         docker logs test-comps-embedding-langchain-mosec-server
         exit 1
     fi
+
+function validate_microservice() {
+    ## query with single text
+    validate_service \
+        '{"text":"What is Deep Learning?"}'
+    
+    ## query with multiple texts
+    validate_service \
+        '{"text":["What is Deep Learning?","How are you?"]}'
+    
+    ## Test OpenAI API, input single text
+    validate_service \
+        '{"input":"What is Deep Learning?"}'
+    
+    ## Test OpenAI API, input multiple texts with parameters
+    validate_service \
+        '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'
 }
 
 function stop_docker() {
diff --git a/tests/embeddings/test_embeddings_tei_langchain.sh b/tests/embeddings/test_embeddings_tei_langchain.sh
index 8c369a116..f8db9824c 100644
--- a/tests/embeddings/test_embeddings_tei_langchain.sh
+++ b/tests/embeddings/test_embeddings_tei_langchain.sh
@@ -30,11 +30,12 @@ function start_service() {
     sleep 3m
 }
 
-function validate_microservice() {
+function validate_service() {
+    local INPUT_DATA="$1"
     tei_service_port=5002
     result=$(http_proxy="" curl http://${ip_address}:$tei_service_port/v1/embeddings \
         -X POST \
-        -d '{"text":"What is Deep Learning?"}' \
+        -d "$INPUT_DATA" \
         -H 'Content-Type: application/json')
     if [[ $result == *"embedding"* ]]; then
         echo "Result correct."
@@ -46,6 +47,24 @@ function validate_microservice() {
     fi
 }
 
+function validate_microservice() {
+    ## query with single text
+    validate_service \
+        '{"text":"What is Deep Learning?"}'
+    
+    ## query with multiple texts
+    validate_service \
+        '{"text":["What is Deep Learning?","How are you?"]}'
+    
+    ## Test OpenAI API, input single text
+    validate_service \
+        '{"input":"What is Deep Learning?"}'
+    
+    ## Test OpenAI API, input multiple texts with parameters
+    validate_service \
+        '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'
+}
+
 function validate_microservice_with_openai() {
     tei_service_port=5002
     python3 ${WORKPATH}/tests/utils/validate_svc_with_openai.py $ip_address $tei_service_port "embedding"
diff --git a/tests/embeddings/test_embeddings_tei_llama_index.sh b/tests/embeddings/test_embeddings_tei_llama_index.sh
index 46bbd150c..15d7d22cc 100644
--- a/tests/embeddings/test_embeddings_tei_llama_index.sh
+++ b/tests/embeddings/test_embeddings_tei_llama_index.sh
@@ -30,14 +30,16 @@ function start_service() {
     sleep 3m
 }
 
-function validate_microservice() {
+function validate_service() {
+    local INPUT_DATA="$1"
+
     tei_service_port=5034
     URL="http://${ip_address}:$tei_service_port/v1/embeddings"
     docker logs test-comps-embedding-tei-llama-index-server >> ${LOG_PATH}/embedding.log
-    HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" -X POST -d '{"text":"What is Deep Learning?"}' -H 'Content-Type: application/json' "$URL")
+    HTTP_STATUS=$(curl -s -o /dev/null -w "%{http_code}" -X POST -d "$INPUT_DATA" -H 'Content-Type: application/json' "$URL")
     if [ "$HTTP_STATUS" -eq 200 ]; then
         echo "[ embedding - llama_index ] HTTP status is 200. Checking content..."
-        local CONTENT=$(curl -s -X POST -d '{"text":"What is Deep Learning?"}' -H 'Content-Type: application/json' "$URL" | tee ${LOG_PATH}/embedding.log)
+        local CONTENT=$(curl -s -X POST -d "$INPUT_DATA" -H 'Content-Type: application/json' "$URL" | tee ${LOG_PATH}/embedding.log)
 
         if echo '"text":"What is Deep Learning?","embedding":\[' | grep -q "$EXPECTED_RESULT"; then
             echo "[ embedding - llama_index ] Content is as expected."
@@ -53,6 +55,24 @@ function validate_microservice() {
     fi
 }
 
+function validate_microservice() {
+    ## query with single text
+    validate_service \
+        '{"text":"What is Deep Learning?"}'
+    
+    ## query with multiple texts
+    validate_service \
+        '{"text":["What is Deep Learning?","How are you?"]}'
+    
+    ## Test OpenAI API, input single text
+    validate_service \
+        '{"input":"What is Deep Learning?"}'
+    
+    ## Test OpenAI API, input multiple texts with parameters
+    validate_service \
+        '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'
+}
+
 function stop_docker() {
     cid=$(docker ps -aq --filter "name=test-comps-embedding-*")
     if [[ ! -z "$cid" ]]; then docker stop $cid && docker rm $cid && sleep 1s; fi

From 66c18681bfe2ef2da58a9ed741a0e887e2aa30d7 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 05:40:37 +0000
Subject: [PATCH 09/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 comps/embeddings/mosec/langchain/embedding_mosec.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/comps/embeddings/mosec/langchain/embedding_mosec.py b/comps/embeddings/mosec/langchain/embedding_mosec.py
index 2f6800277..38e92b5a7 100644
--- a/comps/embeddings/mosec/langchain/embedding_mosec.py
+++ b/comps/embeddings/mosec/langchain/embedding_mosec.py
@@ -24,6 +24,7 @@
     EmbeddingResponse,
     EmbeddingResponseData,
 )
+
 logger = CustomLogger("embedding_mosec")
 logflag = os.getenv("LOGFLAG", False)
 
@@ -81,7 +82,7 @@ async def embedding(
         embed_vector = await get_embeddings(input.input)
         if input.dimensions is not None:
             embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
-        
+
         # for standard openai embedding format
         res = EmbeddingResponse(
             data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))]
@@ -97,11 +98,13 @@ async def embedding(
         logger.info(res)
     return res
 
+
 async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
     texts = [text] if isinstance(text, str) else text
     embed_vector = await embeddings.aembed_documents(texts)
     return embed_vector
 
+
 if __name__ == "__main__":
     MOSEC_EMBEDDING_ENDPOINT = os.environ.get("MOSEC_EMBEDDING_ENDPOINT", "http://127.0.0.1:8080")
     os.environ["OPENAI_API_BASE"] = MOSEC_EMBEDDING_ENDPOINT

From 7e46ac9baa708d0e0600807512d5b5bbdead9ac8 Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 13:42:20 +0800
Subject: [PATCH 10/18] fix ut bug

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 tests/embeddings/test_embeddings_mosec_langchain.sh | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tests/embeddings/test_embeddings_mosec_langchain.sh b/tests/embeddings/test_embeddings_mosec_langchain.sh
index d99d656b0..e2fe1348e 100644
--- a/tests/embeddings/test_embeddings_mosec_langchain.sh
+++ b/tests/embeddings/test_embeddings_mosec_langchain.sh
@@ -57,6 +57,7 @@ function validate_service() {
         docker logs test-comps-embedding-langchain-mosec-server
         exit 1
     fi
+}
 
 function validate_microservice() {
     ## query with single text

From d6f1a1da181c324f5fd6a5e0d52350d93a805b5d Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 06:16:55 +0000
Subject: [PATCH 11/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 tests/embeddings/test_embeddings_mosec_langchain.sh | 6 +++---
 tests/embeddings/test_embeddings_tei_langchain.sh   | 6 +++---
 tests/embeddings/test_embeddings_tei_llama_index.sh | 6 +++---
 3 files changed, 9 insertions(+), 9 deletions(-)

diff --git a/tests/embeddings/test_embeddings_mosec_langchain.sh b/tests/embeddings/test_embeddings_mosec_langchain.sh
index d99d656b0..a2c969de7 100644
--- a/tests/embeddings/test_embeddings_mosec_langchain.sh
+++ b/tests/embeddings/test_embeddings_mosec_langchain.sh
@@ -62,15 +62,15 @@ function validate_microservice() {
     ## query with single text
     validate_service \
         '{"text":"What is Deep Learning?"}'
-    
+
     ## query with multiple texts
     validate_service \
         '{"text":["What is Deep Learning?","How are you?"]}'
-    
+
     ## Test OpenAI API, input single text
     validate_service \
         '{"input":"What is Deep Learning?"}'
-    
+
     ## Test OpenAI API, input multiple texts with parameters
     validate_service \
         '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'
diff --git a/tests/embeddings/test_embeddings_tei_langchain.sh b/tests/embeddings/test_embeddings_tei_langchain.sh
index f8db9824c..0918b025f 100644
--- a/tests/embeddings/test_embeddings_tei_langchain.sh
+++ b/tests/embeddings/test_embeddings_tei_langchain.sh
@@ -51,15 +51,15 @@ function validate_microservice() {
     ## query with single text
     validate_service \
         '{"text":"What is Deep Learning?"}'
-    
+
     ## query with multiple texts
     validate_service \
         '{"text":["What is Deep Learning?","How are you?"]}'
-    
+
     ## Test OpenAI API, input single text
     validate_service \
         '{"input":"What is Deep Learning?"}'
-    
+
     ## Test OpenAI API, input multiple texts with parameters
     validate_service \
         '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'
diff --git a/tests/embeddings/test_embeddings_tei_llama_index.sh b/tests/embeddings/test_embeddings_tei_llama_index.sh
index 15d7d22cc..cc4f769bc 100644
--- a/tests/embeddings/test_embeddings_tei_llama_index.sh
+++ b/tests/embeddings/test_embeddings_tei_llama_index.sh
@@ -59,15 +59,15 @@ function validate_microservice() {
     ## query with single text
     validate_service \
         '{"text":"What is Deep Learning?"}'
-    
+
     ## query with multiple texts
     validate_service \
         '{"text":["What is Deep Learning?","How are you?"]}'
-    
+
     ## Test OpenAI API, input single text
     validate_service \
         '{"input":"What is Deep Learning?"}'
-    
+
     ## Test OpenAI API, input multiple texts with parameters
     validate_service \
         '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'

From f569d346b7094dfb31b8bff98cb69b683fb2622b Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 14:18:58 +0800
Subject: [PATCH 12/18] support embedding predictionguard  openai compatible
 API

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 comps/embeddings/predictionguard/README.md    | 33 +++++++++++--
 .../embedding_predictionguard.py              | 49 ++++++++++++++++---
 .../test_embeddings_predictionguard.sh        | 23 ++++++++-
 3 files changed, 93 insertions(+), 12 deletions(-)

diff --git a/comps/embeddings/predictionguard/README.md b/comps/embeddings/predictionguard/README.md
index 4dccd3354..bec54350c 100644
--- a/comps/embeddings/predictionguard/README.md
+++ b/comps/embeddings/predictionguard/README.md
@@ -31,9 +31,34 @@ docker run -d --name="embedding-predictionguard" -p 6000:6000 -e PREDICTIONGUARD
 
 ## 🚀 Consume Embeddings Service
 
+Use our basic API.
+
 ```bash
-curl localhost:6000/v1/embeddings \
-     -X POST \
-     -d '{"text":"Hello, world!"}' \
-     -H 'Content-Type: application/json'
+## query with single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
+
+## query with multiple texts
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":["Hello, world!","How are you?"]}' \
+  -H 'Content-Type: application/json'
+```
+
+We are also compatible with [OpenAI API](https://platform.openai.com/docs/api-reference/embeddings).
+
+```bash
+## Input single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
+
+## Input multiple texts with parameters
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":["Hello, world!","How are you?"], "dimensions":100}' \
+  -H 'Content-Type: application/json'
 ```
diff --git a/comps/embeddings/predictionguard/embedding_predictionguard.py b/comps/embeddings/predictionguard/embedding_predictionguard.py
index 4ea557f70..e77749fbe 100644
--- a/comps/embeddings/predictionguard/embedding_predictionguard.py
+++ b/comps/embeddings/predictionguard/embedding_predictionguard.py
@@ -4,10 +4,11 @@
 
 import os
 import time
-
+from typing import List, Optional, Union
 from predictionguard import PredictionGuard
 
 from comps import (
+    CustomLogger,
     EmbedDoc,
     ServiceType,
     TextDoc,
@@ -16,6 +17,15 @@
     register_statistics,
     statistics_dict,
 )
+from comps.cores.proto.api_protocol import (
+    ChatCompletionRequest,
+    EmbeddingRequest,
+    EmbeddingResponse,
+    EmbeddingResponseData,
+)
+
+logger = CustomLogger("embedding_predictionguard")
+logflag = os.getenv("LOGFLAG", False)
 
 # Initialize Prediction Guard client
 client = PredictionGuard()
@@ -31,15 +41,42 @@
     output_datatype=EmbedDoc,
 )
 @register_statistics(names=["opea_service@embedding_predictionguard"])
-def embedding(input: TextDoc) -> EmbedDoc:
+async def embedding(
+    input: Union[TextDoc, EmbeddingRequest, ChatCompletionRequest]
+) -> Union[EmbedDoc, EmbeddingResponse, ChatCompletionRequest]:
+    if logflag:
+        logger.info(input)
     start = time.time()
-    response = client.embeddings.create(model=pg_embedding_model_name, input=[{"text": input.text}])
-    embed_vector = response["data"][0]["embedding"]
-    embed_vector = embed_vector[:512]  # Keep only the first 512 elements
-    res = EmbedDoc(text=input.text, embedding=embed_vector)
+
+    if isinstance(input, TextDoc):
+        embed_vector = await get_embeddings(input.text)
+        embedding_res = embed_vector[0] if isinstance(input.text, str) else embed_vector
+        res = EmbedDoc(text=input.text, embedding=embedding_res)
+    else:
+        embed_vector = await get_embeddings(input.input)
+        input.dimensions = input.dimensions if input.dimensions is not None else 512
+        embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
+
+        # for standard openai embedding format
+        res = EmbeddingResponse(
+            data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))]
+        )
+
+        if isinstance(input, ChatCompletionRequest):
+            input.embedding = res
+            # keep
+            res = input
+
     statistics_dict["opea_service@embedding_predictionguard"].append_latency(time.time() - start, None)
+    if logflag:
+        logger.info(res)
     return res
 
+async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
+    texts = [text] if isinstance(text, str) else text
+    response = client.embeddings.create(model=pg_embedding_model_name, input=texts)["data"]
+    embed_vector = [response[i]["embedding"] for i in range(len(response))]
+    return embed_vector
 
 if __name__ == "__main__":
     pg_embedding_model_name = os.getenv("PG_EMBEDDING_MODEL_NAME", "bridgetower-large-itm-mlm-itc")
diff --git a/tests/embeddings/test_embeddings_predictionguard.sh b/tests/embeddings/test_embeddings_predictionguard.sh
index 567b4fc0b..83272908e 100644
--- a/tests/embeddings/test_embeddings_predictionguard.sh
+++ b/tests/embeddings/test_embeddings_predictionguard.sh
@@ -32,11 +32,12 @@ function start_service() {
     sleep 60  # Sleep for 1 minute to allow the service to start
 }
 
-function validate_microservice() {
+function validate_service() {
+    local INPUT_DATA="$1"
     tei_service_port=6000
     result=$(http_proxy="" curl http://${ip_address}:${tei_service_port}/v1/embeddings \
         -X POST \
-        -d '{"text":"What is Deep Learning?"}' \
+        -d "$INPUT_DATA" \
         -H 'Content-Type: application/json')
 
     # Check for a proper response format
@@ -53,6 +54,24 @@ function validate_microservice() {
     fi
 }
 
+function validate_microservice() {
+    ## query with single text
+    validate_service \
+        '{"text":"What is Deep Learning?"}'
+
+    ## query with multiple texts
+    validate_service \
+        '{"text":["What is Deep Learning?","How are you?"]}'
+
+    ## Test OpenAI API, input single text
+    validate_service \
+        '{"input":"What is Deep Learning?"}'
+
+    ## Test OpenAI API, input multiple texts with parameters
+    validate_service \
+        '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'
+}
+
 function stop_docker() {
     cid=$(docker ps -aq --filter "name=test-comps-embedding-pg-*")
     if [[ ! -z "$cid" ]]; then docker stop $cid && docker rm $cid && sleep 1s; fi

From 4a34e44bda7eaaf42c306f04bc1206c4031bb5b7 Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 14:43:58 +0800
Subject: [PATCH 13/18] support embedding multimodal clip OpenAI compatible API

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 comps/cores/proto/docarray.py                 |  2 +-
 comps/embeddings/multimodal_clip/README.md    | 31 ++++++++++--
 .../multimodal_clip/embedding_multimodal.py   | 50 ++++++++++++++++---
 .../test_embeddings_multimodal_clip.sh        | 23 ++++++++-
 4 files changed, 92 insertions(+), 14 deletions(-)

diff --git a/comps/cores/proto/docarray.py b/comps/cores/proto/docarray.py
index 88c67b7a4..50871f7bf 100644
--- a/comps/cores/proto/docarray.py
+++ b/comps/cores/proto/docarray.py
@@ -101,7 +101,7 @@ class EmbedDoc(BaseDoc):
     fetch_k: int = 20
     lambda_mult: float = 0.5
     score_threshold: float = 0.2
-    constraints: Optional[Union[Dict[str, Any], None]] = None
+    constraints: Optional[Union[Dict[str, Any], List[Dict[str, Any]],None]] = None
 
 
 class EmbedMultimodalDoc(EmbedDoc):
diff --git a/comps/embeddings/multimodal_clip/README.md b/comps/embeddings/multimodal_clip/README.md
index c7c74e33d..eb3651495 100644
--- a/comps/embeddings/multimodal_clip/README.md
+++ b/comps/embeddings/multimodal_clip/README.md
@@ -44,9 +44,34 @@ curl http://localhost:6000/v1/health_check\
 
 ### 2.2 Consume Embedding Service
 
+Use our basic API.
+
+```bash
+## query with single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
+
+## query with multiple texts
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"text":["Hello, world!","How are you?"]}' \
+  -H 'Content-Type: application/json'
+```
+
+We are also compatible with [OpenAI API](https://platform.openai.com/docs/api-reference/embeddings).
+
 ```bash
-curl http://localhost:6000/v1/embeddings \
-      -X POST   -d '{"text":"Sample text"}' \
-      -H 'Content-Type: application/json'
+## Input single text
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":"Hello, world!"}' \
+  -H 'Content-Type: application/json'
 
+## Input multiple texts with parameters
+curl http://localhost:6000/v1/embeddings\
+  -X POST \
+  -d '{"input":["Hello, world!","How are you?"], "dimensions":100}' \
+  -H 'Content-Type: application/json'
 ```
diff --git a/comps/embeddings/multimodal_clip/embedding_multimodal.py b/comps/embeddings/multimodal_clip/embedding_multimodal.py
index 334b2c029..a9fa88e5c 100644
--- a/comps/embeddings/multimodal_clip/embedding_multimodal.py
+++ b/comps/embeddings/multimodal_clip/embedding_multimodal.py
@@ -3,11 +3,12 @@
 
 import datetime
 import time
-
+from typing import List, Optional, Union
 from dateparser.search import search_dates
 from embeddings_clip import vCLIP
 
 from comps import (
+    CustomLogger,
     EmbedDoc,
     ServiceType,
     TextDoc,
@@ -16,7 +17,14 @@
     register_statistics,
     statistics_dict,
 )
-
+from comps.cores.proto.api_protocol import (
+    ChatCompletionRequest,
+    EmbeddingRequest,
+    EmbeddingResponse,
+    EmbeddingResponseData,
+)
+logger = CustomLogger("embedding_multimodal")
+logflag = os.getenv("LOGFLAG", False)
 
 def filtler_dates(prompt):
 
@@ -64,20 +72,46 @@ def filtler_dates(prompt):
     output_datatype=EmbedDoc,
 )
 @register_statistics(names=["opea_service@embedding_multimodal"])
-def embedding(input: TextDoc) -> EmbedDoc:
+async def embedding(
+    input: Union[TextDoc, EmbeddingRequest, ChatCompletionRequest]
+) -> Union[EmbedDoc, EmbeddingResponse, ChatCompletionRequest]:
+    if logflag:
+        logger.info(input)
     start = time.time()
 
     if isinstance(input, TextDoc):
-        # Handle text input
-        embed_vector = embeddings.embed_query(input.text).tolist()[0]
-        res = EmbedDoc(text=input.text, embedding=embed_vector, constraints=filtler_dates(input.text))
-
+        embed_vector = await get_embeddings(input.text)
+        if isinstance(input.text, str):
+            embedding_res = embed_vector[0]
+            constraints_res = filtler_dates(input.text)
+        else:
+            embedding_res = embed_vector
+            constraints_res = [filtler_dates(input.text[i]) for i in range(len(input.text))]
+        res = EmbedDoc(text=input.text, embedding=embedding_res, constraints=constraints_res)
     else:
-        raise ValueError("Invalid input type")
+        embed_vector = await get_embeddings(input.input)
+        if input.dimensions is not None:
+            embed_vector = [embed_vector[i][: input.dimensions] for i in range(len(embed_vector))]
+
+        # for standard openai embedding format
+        res = EmbeddingResponse(
+            data=[EmbeddingResponseData(index=i, embedding=embed_vector[i]) for i in range(len(embed_vector))]
+        )
+
+        if isinstance(input, ChatCompletionRequest):
+            input.embedding = res
+            # keep
+            res = input
 
     statistics_dict["opea_service@embedding_multimodal"].append_latency(time.time() - start, None)
+    if logflag:
+        logger.info(res)
     return res
 
+async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
+    texts = [text] if isinstance(text, str) else text
+    embed_vector = embeddings.embed_query(texts).tolist()
+    return embed_vector
 
 if __name__ == "__main__":
     embeddings = vCLIP({"model_name": "openai/clip-vit-base-patch32", "num_frm": 4})
diff --git a/tests/embeddings/test_embeddings_multimodal_clip.sh b/tests/embeddings/test_embeddings_multimodal_clip.sh
index c3d0bccde..802d5350a 100644
--- a/tests/embeddings/test_embeddings_multimodal_clip.sh
+++ b/tests/embeddings/test_embeddings_multimodal_clip.sh
@@ -24,11 +24,12 @@ function start_service() {
     sleep 3m
 }
 
-function validate_microservice() {
+function validate_service() {
+    local INPUT_DATA="$1"
     service_port=5038
     result=$(http_proxy="" curl http://${ip_address}:$service_port/v1/embeddings \
         -X POST \
-        -d '{"text":"how many cars are in this image?"}' \
+        -d "$INPUT_DATA" \
         -H 'Content-Type: application/json')
     if [[ $result == *"embedding"* ]]; then
         echo "Result correct."
@@ -39,6 +40,24 @@ function validate_microservice() {
     fi
 }
 
+function validate_microservice() {
+    ## query with single text
+    validate_service \
+        '{"text":"What is Deep Learning?"}'
+
+    ## query with multiple texts
+    validate_service \
+        '{"text":["What is Deep Learning?","How are you?"]}'
+
+    ## Test OpenAI API, input single text
+    validate_service \
+        '{"input":"What is Deep Learning?"}'
+
+    ## Test OpenAI API, input multiple texts with parameters
+    validate_service \
+        '{"input":["What is Deep Learning?","How are you?"], "dimensions":100}'
+}
+
 function stop_docker() {
     cid=$(docker ps -aq --filter "name=test-embedding-multimodal-server-*")
     if [[ ! -z "$cid" ]]; then docker stop $cid && docker rm $cid && sleep 1s; fi

From 792e419e6c06406521b32865e6339de6daea6ff2 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 06:59:18 +0000
Subject: [PATCH 14/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 comps/embeddings/predictionguard/embedding_predictionguard.py | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/comps/embeddings/predictionguard/embedding_predictionguard.py b/comps/embeddings/predictionguard/embedding_predictionguard.py
index e77749fbe..793f35d2d 100644
--- a/comps/embeddings/predictionguard/embedding_predictionguard.py
+++ b/comps/embeddings/predictionguard/embedding_predictionguard.py
@@ -5,6 +5,7 @@
 import os
 import time
 from typing import List, Optional, Union
+
 from predictionguard import PredictionGuard
 
 from comps import (
@@ -72,12 +73,14 @@ async def embedding(
         logger.info(res)
     return res
 
+
 async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
     texts = [text] if isinstance(text, str) else text
     response = client.embeddings.create(model=pg_embedding_model_name, input=texts)["data"]
     embed_vector = [response[i]["embedding"] for i in range(len(response))]
     return embed_vector
 
+
 if __name__ == "__main__":
     pg_embedding_model_name = os.getenv("PG_EMBEDDING_MODEL_NAME", "bridgetower-large-itm-mlm-itc")
     print("Prediction Guard Embedding initialized.")

From 3e4be8032a9a04dd6441a891710e930a17d18261 Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 14:52:36 +0800
Subject: [PATCH 15/18] fix bug

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 comps/embeddings/multimodal_clip/embedding_multimodal.py      | 2 +-
 comps/embeddings/predictionguard/embedding_predictionguard.py | 1 +
 2 files changed, 2 insertions(+), 1 deletion(-)

diff --git a/comps/embeddings/multimodal_clip/embedding_multimodal.py b/comps/embeddings/multimodal_clip/embedding_multimodal.py
index a9fa88e5c..2401326a0 100644
--- a/comps/embeddings/multimodal_clip/embedding_multimodal.py
+++ b/comps/embeddings/multimodal_clip/embedding_multimodal.py
@@ -2,7 +2,7 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import datetime
-import time
+import time,os
 from typing import List, Optional, Union
 from dateparser.search import search_dates
 from embeddings_clip import vCLIP
diff --git a/comps/embeddings/predictionguard/embedding_predictionguard.py b/comps/embeddings/predictionguard/embedding_predictionguard.py
index 793f35d2d..f5274e3b5 100644
--- a/comps/embeddings/predictionguard/embedding_predictionguard.py
+++ b/comps/embeddings/predictionguard/embedding_predictionguard.py
@@ -76,6 +76,7 @@ async def embedding(
 
 async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
     texts = [text] if isinstance(text, str) else text
+    texts = [{"text": texts[i]} for i in range(len(texts))]
     response = client.embeddings.create(model=pg_embedding_model_name, input=texts)["data"]
     embed_vector = [response[i]["embedding"] for i in range(len(response))]
     return embed_vector

From e9a6159322713f950584da673b7eb93633e5d64e Mon Sep 17 00:00:00 2001
From: Xinyao Wang <xinyao.wang@intel.com>
Date: Wed, 13 Nov 2024 14:55:26 +0800
Subject: [PATCH 16/18] enable debug mode for embedding UT

Signed-off-by: Xinyao Wang <xinyao.wang@intel.com>
---
 tests/embeddings/test_embeddings_mosec_langchain.sh | 2 +-
 tests/embeddings/test_embeddings_multimodal_clip.sh | 2 +-
 tests/embeddings/test_embeddings_predictionguard.sh | 2 +-
 tests/embeddings/test_embeddings_tei_langchain.sh   | 2 +-
 tests/embeddings/test_embeddings_tei_llama_index.sh | 2 +-
 5 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/tests/embeddings/test_embeddings_mosec_langchain.sh b/tests/embeddings/test_embeddings_mosec_langchain.sh
index 22ece5395..0c7b1bc3c 100644
--- a/tests/embeddings/test_embeddings_mosec_langchain.sh
+++ b/tests/embeddings/test_embeddings_mosec_langchain.sh
@@ -38,7 +38,7 @@ function start_service() {
     docker run -d --name="test-comps-embedding-langchain-mosec-endpoint" -p $mosec_endpoint:8000  opea/embedding-langchain-mosec-endpoint:comps
     export MOSEC_EMBEDDING_ENDPOINT="http://${ip_address}:${mosec_endpoint}"
     mosec_service_port=5002
-    docker run -d --name="test-comps-embedding-langchain-mosec-server" -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p ${mosec_service_port}:6000 --ipc=host -e MOSEC_EMBEDDING_ENDPOINT=$MOSEC_EMBEDDING_ENDPOINT  opea/embedding-langchain-mosec:comps
+    docker run -d --name="test-comps-embedding-langchain-mosec-server" -e LOGFLAG=True -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p ${mosec_service_port}:6000 --ipc=host -e MOSEC_EMBEDDING_ENDPOINT=$MOSEC_EMBEDDING_ENDPOINT  opea/embedding-langchain-mosec:comps
     sleep 3m
 }
 
diff --git a/tests/embeddings/test_embeddings_multimodal_clip.sh b/tests/embeddings/test_embeddings_multimodal_clip.sh
index 802d5350a..770f2dc3d 100644
--- a/tests/embeddings/test_embeddings_multimodal_clip.sh
+++ b/tests/embeddings/test_embeddings_multimodal_clip.sh
@@ -20,7 +20,7 @@ function build_docker_images() {
 }
 
 function start_service() {
-    docker run -d --name="test-embedding-multimodal-server" -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p 5038:6000 --ipc=host  opea/embedding-multimodal:comps
+    docker run -d --name="test-embedding-multimodal-server" -e LOGFLAG=True -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p 5038:6000 --ipc=host  opea/embedding-multimodal:comps
     sleep 3m
 }
 
diff --git a/tests/embeddings/test_embeddings_predictionguard.sh b/tests/embeddings/test_embeddings_predictionguard.sh
index 83272908e..a6727b6bf 100644
--- a/tests/embeddings/test_embeddings_predictionguard.sh
+++ b/tests/embeddings/test_embeddings_predictionguard.sh
@@ -26,7 +26,7 @@ function start_service() {
     tei_service_port=6000
     unset http_proxy
     docker run -d --name=test-comps-embedding-pg-server \
-    -e http_proxy= -e https_proxy= \
+    -e LOGFLAG=True -e http_proxy= -e https_proxy= \
     -e PREDICTIONGUARD_API_KEY=${PREDICTIONGUARD_API_KEY} \
     -p 6000:6000 --ipc=host opea/embedding-pg:comps
     sleep 60  # Sleep for 1 minute to allow the service to start
diff --git a/tests/embeddings/test_embeddings_tei_langchain.sh b/tests/embeddings/test_embeddings_tei_langchain.sh
index 0918b025f..df2642cf1 100644
--- a/tests/embeddings/test_embeddings_tei_langchain.sh
+++ b/tests/embeddings/test_embeddings_tei_langchain.sh
@@ -26,7 +26,7 @@ function start_service() {
     docker run -d --name="test-comps-embedding-tei-endpoint" -p $tei_endpoint:80 -v ./data:/data --pull always ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 --model-id $model
     export TEI_EMBEDDING_ENDPOINT="http://${ip_address}:${tei_endpoint}"
     tei_service_port=5002
-    docker run -d --name="test-comps-embedding-tei-server" -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p ${tei_service_port}:6000 --ipc=host -e TEI_EMBEDDING_ENDPOINT=$TEI_EMBEDDING_ENDPOINT  opea/embedding-tei:comps
+    docker run -d --name="test-comps-embedding-tei-server" -e LOGFLAG=True -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p ${tei_service_port}:6000 --ipc=host -e TEI_EMBEDDING_ENDPOINT=$TEI_EMBEDDING_ENDPOINT  opea/embedding-tei:comps
     sleep 3m
 }
 
diff --git a/tests/embeddings/test_embeddings_tei_llama_index.sh b/tests/embeddings/test_embeddings_tei_llama_index.sh
index cc4f769bc..e1d04ab5e 100644
--- a/tests/embeddings/test_embeddings_tei_llama_index.sh
+++ b/tests/embeddings/test_embeddings_tei_llama_index.sh
@@ -26,7 +26,7 @@ function start_service() {
     docker run -d --name="test-comps-embedding-tei-llama-index-endpoint" -p $tei_endpoint:80 -v ./data:/data -e http_proxy=$http_proxy -e https_proxy=$https_proxy --pull always ghcr.io/huggingface/text-embeddings-inference:cpu-1.5 --model-id $model
     export TEI_EMBEDDING_ENDPOINT="http://${ip_address}:${tei_endpoint}"
     tei_service_port=5034
-    docker run -d --name="test-comps-embedding-tei-llama-index-server" -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p ${tei_service_port}:6000 --ipc=host -e TEI_EMBEDDING_ENDPOINT=$TEI_EMBEDDING_ENDPOINT  opea/embedding-tei-llama-index:comps
+    docker run -d --name="test-comps-embedding-tei-llama-index-server" -e LOGFLAG=True  -e http_proxy=$http_proxy -e https_proxy=$https_proxy -p ${tei_service_port}:6000 --ipc=host -e TEI_EMBEDDING_ENDPOINT=$TEI_EMBEDDING_ENDPOINT  opea/embedding-tei-llama-index:comps
     sleep 3m
 }
 

From d2e5df1e4d0d1ed08d43b3b1d897698a626fa291 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 07:28:09 +0000
Subject: [PATCH 17/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 comps/cores/proto/docarray.py                            | 2 +-
 comps/embeddings/multimodal_clip/embedding_multimodal.py | 5 +++++
 2 files changed, 6 insertions(+), 1 deletion(-)

diff --git a/comps/cores/proto/docarray.py b/comps/cores/proto/docarray.py
index 50871f7bf..79f961ca2 100644
--- a/comps/cores/proto/docarray.py
+++ b/comps/cores/proto/docarray.py
@@ -101,7 +101,7 @@ class EmbedDoc(BaseDoc):
     fetch_k: int = 20
     lambda_mult: float = 0.5
     score_threshold: float = 0.2
-    constraints: Optional[Union[Dict[str, Any], List[Dict[str, Any]],None]] = None
+    constraints: Optional[Union[Dict[str, Any], List[Dict[str, Any]], None]] = None
 
 
 class EmbedMultimodalDoc(EmbedDoc):
diff --git a/comps/embeddings/multimodal_clip/embedding_multimodal.py b/comps/embeddings/multimodal_clip/embedding_multimodal.py
index a9fa88e5c..c791f1471 100644
--- a/comps/embeddings/multimodal_clip/embedding_multimodal.py
+++ b/comps/embeddings/multimodal_clip/embedding_multimodal.py
@@ -4,6 +4,7 @@
 import datetime
 import time
 from typing import List, Optional, Union
+
 from dateparser.search import search_dates
 from embeddings_clip import vCLIP
 
@@ -23,9 +24,11 @@
     EmbeddingResponse,
     EmbeddingResponseData,
 )
+
 logger = CustomLogger("embedding_multimodal")
 logflag = os.getenv("LOGFLAG", False)
 
+
 def filtler_dates(prompt):
 
     base_date = datetime.datetime.today()
@@ -108,11 +111,13 @@ async def embedding(
         logger.info(res)
     return res
 
+
 async def get_embeddings(text: Union[str, List[str]]) -> List[List[float]]:
     texts = [text] if isinstance(text, str) else text
     embed_vector = embeddings.embed_query(texts).tolist()
     return embed_vector
 
+
 if __name__ == "__main__":
     embeddings = vCLIP({"model_name": "openai/clip-vit-base-patch32", "num_frm": 4})
     opea_microservices["opea_service@embedding_multimodal"].start()

From e429ca01d8da2bdb9871a350137ac276cb1de439 Mon Sep 17 00:00:00 2001
From: "pre-commit-ci[bot]"
 <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Date: Wed, 13 Nov 2024 07:35:47 +0000
Subject: [PATCH 18/18] [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci
---
 comps/embeddings/multimodal_clip/embedding_multimodal.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/comps/embeddings/multimodal_clip/embedding_multimodal.py b/comps/embeddings/multimodal_clip/embedding_multimodal.py
index aafee3986..888278e7a 100644
--- a/comps/embeddings/multimodal_clip/embedding_multimodal.py
+++ b/comps/embeddings/multimodal_clip/embedding_multimodal.py
@@ -2,7 +2,8 @@
 # SPDX-License-Identifier: Apache-2.0
 
 import datetime
-import time,os
+import os
+import time
 from typing import List, Optional, Union
 
 from dateparser.search import search_dates