From f28bf3f13d4f024c32d50f67f311f3022d233e26 Mon Sep 17 00:00:00 2001
From: Andrei Betlen <abetlen@gmail.com>
Date: Sat, 1 Apr 2023 15:12:25 -0400
Subject: [PATCH] Bugfix: enable embeddings for fastapi server

---
 examples/fastapi_server.py | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/examples/fastapi_server.py b/examples/fastapi_server.py
index 7be2b05..04dcaab 100644
--- a/examples/fastapi_server.py
+++ b/examples/fastapi_server.py
@@ -31,6 +31,7 @@ llama = llama_cpp.Llama(
     settings.model,
     f16_kv=True,
     use_mlock=True,
+    embedding=True,
     n_threads=6,
     n_batch=2048,
 )
@@ -93,4 +94,6 @@ CreateEmbeddingResponse = create_model_from_typeddict(llama_cpp.Embedding)
     response_model=CreateEmbeddingResponse,
 )
 def create_embedding(request: CreateEmbeddingRequest):
-    return llama.create_embedding(**request.dict())
+    # print(request)
+    # return llama.create_embedding(**request.dict(exclude={"model", "user"}))
+    return llama.create_embedding(request.input)