From 512769e84105a24fd7118ba18b71805ab94c28d3 Mon Sep 17 00:00:00 2001
From: Krrish Dholakia <krrishdholakia@gmail.com>
Date: Tue, 3 Oct 2023 07:10:05 -0700
Subject: [PATCH] support n param for hf

---
 litellm/__pycache__/utils.cpython-311.pyc | Bin 139358 -> 139395 bytes
 litellm/llms/huggingface_restapi.py       |   2 --
 litellm/tests/test_streaming.py           |   2 +-
 litellm/utils.py                          |   4 +++-
 pyproject.toml                            |   2 +-
 5 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/litellm/__pycache__/utils.cpython-311.pyc b/litellm/__pycache__/utils.cpython-311.pyc
index 6459cd2d8adb31965cd5580ce5ef0f1155c12788..9cef80ae0463a4dfda4c047a3e04743ff45a5ff0 100644
GIT binary patch
delta 654
zcmZXPSxD4z9K`qY+rO@xtC>gTQJF24M;8@E6j9)Vl}H)2VDyk1WkrXA6xb=1L~hcb
z+LCq?=`Je!D_`u=A~BXUH4_7mL@1?)woDSj%6)Da=4EDJrne!a<Yh?st+22Org2P}
zp0LASULAg;0*6sk(E{i6lB-9^((&AxG+G?)Ox&Xa!Gi3aCg)6jl4qLU<=JM{3WAEY
z;%H}*vCw-W?85zcRfyrcmuY6R5&dwf+2Ux84^XE%POmrYdimp=2%F!8HLWWpwKV&3
zW}7Y9=zlz5qr+&dZ?X_+lr=fcY&CxPbj3mAbn`OEGwNH^M9Pd;J+q=nH$MB5CQ~9-
z3!yQjNlXHL!HA3`F$T|N*iJG$mAGVfE1Koh9EPD=4ySSkhJ<u}oB1wd^Qj9WQe+7W
zf><qs%gImVm~6}?)=IwYb@D8hZ<2D98ydgl-AVQb5i6@o*`-(^x6fiIvfNwC7{uYu
zaB+xS_X`)V*d`~=)2`Sqn=X)QrcjPuBu?XuR9s<1aZSpv(nj#zP?M=}pF6vfd2-Q9
z4b;loN>-@ghJf=R;LH}Iigk*1`FkB742bV08G$1aw~5lm2iotmK_gU(9&sm{8IssY
z#{}x!Bk$;By}WJZlA=NK-t!Zh#PNYlD<kfzUcM3`u09-!PWjT$TMLV%W{3#Oc&*XT
z8w*9^8pW&eK<19I1)j*MKX~DFtzAun4)+?AhmZ0Jm80m9#hO}Y;g1{-Rk<1~rF)`s
T+NK`Ri1(ZRQ>Qo*RIv6B5ZCDu

delta 659
zcmZXPSxA&&6o%*dW^)_UEO()@tR%uNq_8LmTT}>IsLU=5CM*M`1uW3OL`I5?`l+AH
znZ!^_h0?D$iIFLUkriSaWhqu<8KI&=P*zU29bV4G`|uvlTh<vc@E{<x!Q~3F%#2wT
zxBv6Gme9H?S7xBa;&FK*JyD)$)9hSBobIn`fO4a(x`d4e-W%cL#i8zbhm=2gWOJUy
zJ>LlTCK$O_<Bdu0dY$eoc40R<n_~1}(<UqFdhyMbR_Bi1IKPaP+iQ($t=n~g9txpP
zk92f93Dd3JZYz6qRF9!3)J;!U!9M-znVQ8JJ^Vf~f@B>$z95oPS?(lwE=wgXo<Rq9
zBsP(0XqV_kBte_3OCn#<BaKU#fVbjH#skBWw2YrtMrA<?ZFa)sT`DPd)`@F12j@{D
zMLEO-k|*cg+>7D6Y&l7tjVYNtMWLN2aaV9aktR=1qZL_Z=~=$8)6B2rIN9dMO1ju9
z_2=kQ<jRQ)BwHPagDsY$Z_SokBBkRZf2@>C$7QCJ!|Nx-ezMrSQp0SSsK&r?-ESBC
zvqk%82Q7s)tWn&T=30sry|S>5$1o(L^=$C>O1(~mZRX#MX13Z0lq<KmF^2@%@Q@cl
zh#A|<D>7xgmrMRL70;Q3hf?yAj6f#L?g73LDteIJiWd^{fhSH@$eXW(IcH)cV?1(l
zO!N<YHkxJIG`rxgH2%g1U3W560`!@AC@-(YuGDr#zhv7~mXlvn7p!t@tQNwQJ0!Zm
QM%<X?znv>3@ycQQ3n1I;s{jB1

diff --git a/litellm/llms/huggingface_restapi.py b/litellm/llms/huggingface_restapi.py
index 90abdb2d2..0739d1e92 100644
--- a/litellm/llms/huggingface_restapi.py
+++ b/litellm/llms/huggingface_restapi.py
@@ -117,7 +117,6 @@ def completion(
         inference_params = copy.deepcopy(optional_params)
         inference_params.pop("details")
         inference_params.pop("return_full_text")
-        inference_params.pop("task")
         past_user_inputs = []
         generated_responses = []
         text = ""
@@ -181,7 +180,6 @@ def completion(
         inference_params = copy.deepcopy(optional_params)
         inference_params.pop("details")
         inference_params.pop("return_full_text")
-        inference_params.pop("task")
         data = {
             "inputs": prompt,
             "parameters": inference_params,
diff --git a/litellm/tests/test_streaming.py b/litellm/tests/test_streaming.py
index f016ae97d..1bf8eadf9 100644
--- a/litellm/tests/test_streaming.py
+++ b/litellm/tests/test_streaming.py
@@ -351,7 +351,7 @@ def test_completion_cohere_stream_bad_key():
 #             },
 #         ]
 #         response = completion(
-#             model="huggingface/mistralai/Mistral-7B-Instruct-v0.1", messages=messages, api_base="https://n9ox93a8sv5ihsow.us-east-1.aws.endpoints.huggingface.cloud", stream=True, max_tokens=1000
+#             model="huggingface/mistralai/Mistral-7B-Instruct-v0.1", messages=messages, api_base="https://n9ox93a8sv5ihsow.us-east-1.aws.endpoints.huggingface.cloud", stream=True, max_tokens=1000, n=1
 #         )
 #         complete_response = ""
 #         # Add any assertions here to check the response
diff --git a/litellm/utils.py b/litellm/utils.py
index b7995863d..122659829 100644
--- a/litellm/utils.py
+++ b/litellm/utils.py
@@ -1039,7 +1039,7 @@ def get_optional_params(  # use the openai defaults
             optional_params["stop_sequences"] = stop
     elif custom_llm_provider == "huggingface":
         ## check if unsupported param passed in 
-        supported_params = ["stream", "temperature", "max_tokens", "top_p", "stop",]
+        supported_params = ["stream", "temperature", "max_tokens", "top_p", "stop", "n"]
         _check_valid_arg(supported_params=supported_params)
         
         if temperature:
@@ -1055,6 +1055,8 @@ def get_optional_params(  # use the openai defaults
             optional_params["stop"] = stop
         if max_tokens:
             optional_params["max_new_tokens"] = max_tokens
+        if n: 
+            optional_params["best_of"] = n
         if presence_penalty:
             optional_params["repetition_penalty"] = presence_penalty
     elif custom_llm_provider == "together_ai":
diff --git a/pyproject.toml b/pyproject.toml
index eedbb81a7..1bc0aa326 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "litellm"
-version = "0.1.815"
+version = "0.1.816"
 description = "Library to easily interface with LLM API providers"
 authors = ["BerriAI"]
 license = "MIT License"