Update apis/chat_api.py

apis/chat_api.py

@@ -48,10 +48,10 @@ class EmbeddingsAPIInference:
         return requests.post(API_URL, headers=headers, json=payload).json()
 
 
-class SentenceTransformer(SentenceTransformer):
+class SentenceTransformerLocal(SentenceTransformer):
     def encode(self, **kwargs):
         kwargs.pop("api_key", None)
-        return super().encode(**kwargs)
+        return super().encode(**kwargs).tolist()
         
         
         
@@ -66,9 +66,9 @@ class ChatAPIApp:
         )
         self.setup_routes()
         self.embeddings = {
-            "mxbai-embed-large":SentenceTransformer("mixedbread-ai/mxbai-embed-large-v1"),
-            "nomic-embed-text": SentenceTransformer("nomic-ai/nomic-embed-text-v1.5", trust_remote_code=True),
-            "multilingual-e5-large-instruct":SentenceTransformer("intfloat/multilingual-e5-large-instruct"),
+            "mxbai-embed-large":SentenceTransformerLocal("mixedbread-ai/mxbai-embed-large-v1"),
+            "nomic-embed-text": SentenceTransformerLocal("nomic-ai/nomic-embed-text-v1.5", trust_remote_code=True),
+            "multilingual-e5-large-instruct":SentenceTransformerLocal("intfloat/multilingual-e5-large-instruct"),
             "intfloat/multilingual-e5-large-instruct":EmbeddingsAPIInference("intfloat/multilingual-e5-large-instruct"),
             "mixedbread-ai/mxbai-embed-large-v1":EmbeddingsAPIInference("mixedbread-ai/mxbai-embed-large-v1")
         }
@@ -360,7 +360,7 @@ class ChatAPIApp:
             return {
                 "object":"list", 
                 "data":[
-                    {"object": "embedding", "index": i, "embedding": emb} for i,emb in enumerate(embeddings.tolist())
+                    {"object": "embedding", "index": i, "embedding": emb} for i,emb in enumerate(embeddings)#.tolist())
                 ], 
                 "model": model, 
                 "usage":{},
@@ -373,7 +373,7 @@ class ChatAPIApp:
             model = request.model
             model_kwargs = request.options
             embeddings = self.embeddings[model].encode(request.prompt, api_key=api_key)#, **model_kwargs)
-            return {"embedding": embeddings.tolist()}
+            return {"embedding": embeddings}#.tolist()}
         except ValueError as e:
             raise HTTPException(status_code=400, detail=str(e))