From d3340311084ac9512945fc088ff55e780f676a59 Mon Sep 17 00:00:00 2001
From: Krrish Dholakia <krrishdholakia@gmail.com>
Date: Wed, 27 Sep 2023 21:04:15 -0700
Subject: [PATCH] adding support for completions endpoint in proxy

---
 litellm/__pycache__/main.cpython-311.pyc | Bin 50033 -> 50660 bytes
 litellm/main.py                          |  20 +++++++++++++++++++-
 litellm/proxy/proxy_cli.py               |   3 ++-
 litellm/proxy/proxy_server.py            |  21 ++++++++++++++++++++-
 pyproject.toml                           |   2 +-
 5 files changed, 42 insertions(+), 4 deletions(-)

diff --git a/litellm/__pycache__/main.cpython-311.pyc b/litellm/__pycache__/main.cpython-311.pyc
index 64c0ff698f33e7cb95496f33d205e75af1d5befa..3a1ff5970bd702327e271354fbed505d3105bc1e 100644
GIT binary patch
delta 947
zcmZWnO-vI(6rOE&`={Mv3;jXB?FC#!Ot1-hz(i=o_)j!m`P;SBY&*^D1_2>WGy#mq
zRp|i@UKBy4qyb5Y(WqAqCVOal@}S1!q8>2u;G0EK(3hDn^S<xRdvAW$$L*hQ+xf>{
zuLpr`Y*fnG-;4a)xsV%NAIu+@8HA>g#Ede7h)n6<=K5O3?4x#xi_jgGNe=LxWgk4(
z4$rNBTnyGoyUL`ryS#>QE5I*#%cO+1=&ZzVqkYqCf&r;w8@FfLF~*Lv7m>zowS<u5
ze*~S+!`!qRj2JjarokyW)L~VVQ%`IqmS9CDXMiM~6eKG9G<itt?#HQ&B5O&UO_Q@!
zJ3-+T1sjFa6s(=+p+?DjC7DjD1KmXKRdHHm$$hF@l1ssf48eUWNkJre0D|b)Eakv`
zSLJ@~-vTsbBmw1nRawYpdR0|`1gVTBB-0uev;kS@+1X^eN08G}27@jf{7>AU!jh~s
z{Y}!=Fd!=m7Or8UNDX3-XeXCwW39B~5UfOww9)l33Y~{3b^w$-L{>96t;*dkD!ple
z8nJYRCZKQ^1ogYVGT)vJ)aJCcK=aHX0E|i3Cx3M=cBi=*Y$%5I&A5$FW3j3>moTd8
zO$UlZXCg-U!0d4&+&nvMgpaO>MmSysS~Ma}vt35yP~K@ojudNzVpN<Jjc6?2Wkg$7
zd`9%7>E^t=i742eVcx7j!MfOH=DmE-V<Lb_k6DK*s@GhRFMQZUsJg?p4u3pu`)MIC
z?zpW``Sc#cX>gE*!ossghIw5mJ)dK^>%hD#+{*hn8^QYeQb5ly@tpOb`unAjm%c1q
z8~s0)eu2Sy`|=2r)n6`4+ytl}3iU5;vO$`fJfvYXWr<aiszxUCSF5$291t*&I$rp?
T$~yfaCT=s)HjL@N-VFT#*hB#q

delta 329
zcmaFT&HS;AnQu8SFBbz4%-$-T+VOKE-{Do0*R^u;6#<2&Go&!qFqSbC8BKn%Dv2YQ
zA%!8BL6dc|*%}2mO{Ss{Mh1o=bD%m+<|11lRb&Su+<`<fGmubFP$)76(zp0aQY%X0
zlk;;6a#Bk&^Ye;wfEuT=0+mHhp0n0TieGYu$pwC`1!h13ggCT5a58X6U0|2oJaOGL
zM%DtL^0}LJHfS+2uG<{GaTOzLIZ$NNX12}ztdozfk(jKtgOC3fZ(3$vW^qP*QEFmw
ze%@r;9pb#zKxLY&MJXU>RZdRa5x`hEdDD(6#@flcJ5yL&fwFrxPutndA_lUdXdy@$
t$mClrIhn;JMXi(d_sMci0}22ULvi!wuzf7-{H%=H9~j_-z~r2x6#&jbVebF{

diff --git a/litellm/main.py b/litellm/main.py
index 20e554c7687..160f786f5f5 100644
--- a/litellm/main.py
+++ b/litellm/main.py
@@ -1431,7 +1431,25 @@ def text_completion(*args, **kwargs):
         messages = [{"role": "system", "content": kwargs["prompt"]}]
         kwargs["messages"] = messages
         kwargs.pop("prompt")
-        return completion(*args, **kwargs)
+        response = completion(*args, **kwargs) # assume the response is the openai response object 
+        response_2 = {
+            "id": response["id"],
+            "object": "text_completion",
+            "created": response["created"],
+            "model": response["model"],
+            "choices": [
+            {
+                "text": response["choices"][0]["message"]["content"],
+                "index": response["choices"][0]["index"],
+                "logprobs": None,
+                "finish_reason": response["choices"][0]["finish_reason"]
+            }
+            ],
+            "usage": response["usage"]
+        }
+        return response_2
+    else:
+        raise ValueError("please pass prompt into the `text_completion` endpoint - `text_completion(model, prompt='hello world')`")
 
 ##### Moderation #######################
 def moderation(input: str, api_key: Optional[str]=None):
diff --git a/litellm/proxy/proxy_cli.py b/litellm/proxy/proxy_cli.py
index 75bad1b6fb2..82856e45898 100644
--- a/litellm/proxy/proxy_cli.py
+++ b/litellm/proxy/proxy_cli.py
@@ -7,7 +7,8 @@ load_dotenv()
 @click.option('--api_base', default=None, help='API base URL.')
 @click.option('--model', required=True, help='The model name to pass to litellm expects') 
 def run_server(port, api_base, model):
-    from .proxy_server import app, initialize
+    # from .proxy_server import app, initialize
+    from proxy_server import app, initialize
     initialize(model, api_base)
     try:
         import uvicorn
diff --git a/litellm/proxy/proxy_server.py b/litellm/proxy/proxy_server.py
index 4535775d216..a1934242705 100644
--- a/litellm/proxy/proxy_server.py
+++ b/litellm/proxy/proxy_server.py
@@ -1,4 +1,10 @@
+import sys, os
+sys.path.insert(
+    0, os.path.abspath("../..")
+)  # Adds the parent directory to the system path
+
 import litellm
+print(litellm.__file__)
 from fastapi import FastAPI, Request
 from fastapi.responses import StreamingResponse
 import json
@@ -25,8 +31,21 @@ def model_list():
         object="list",
     )
 
-@app.post("/chat/completions")
+@app.post("/completions")
 async def completion(request: Request):
+    data = await request.json()
+    if (user_model is None):
+        raise ValueError("Proxy model needs to be set")
+    data["model"] = user_model
+    if user_api_base:
+        data["api_base"] = user_api_base
+    response = litellm.text_completion(**data)
+    if 'stream' in data and data['stream'] == True: # use generate_responses to stream responses
+            return StreamingResponse(data_generator(response), media_type='text/event-stream')
+    return response
+
+@app.post("/chat/completions")
+async def chat_completion(request: Request):
     data = await request.json()
     if (user_model is None):
         raise ValueError("Proxy model needs to be set")
diff --git a/pyproject.toml b/pyproject.toml
index 3ffbfd1a760..71d5cdf42b6 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "litellm"
-version = "0.1.789"
+version = "0.1.790"
 description = "Library to easily interface with LLM API providers"
 authors = ["BerriAI"]
 license = "MIT License"