phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 18:54:30 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

bump version

Browse source
This commit is contained in:
ishaan-jaff 2023-09-15 14:14:47 -07:00
parent 9e808c0c1c
commit 5ec1fc5048
3 changed files with 27 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

13
litellm/utils.py
View file

@ -885,8 +885,8 @@ def get_optional_params( # use the openai defaults
if stop != None:
optional_params["stop"] = stop #TG AI expects a list, example ["\n\n\n\n","<|endoftext|>"]
elif (
model == "chat-bison"
): # chat-bison has diff args from chat-bison@001 ty Google
model in litellm.vertex_chat_models or model in litellm.vertex_code_chat_models
): # chat-bison has diff args from chat-bison@001, ty Google :)
if temperature != 1:
optional_params["temperature"] = temperature
if top_p != 1:
@ -900,6 +900,12 @@ def get_optional_params( # use the openai defaults
optional_params["temperature"] = temperature
optional_params["top_p"] = top_p
optional_params["top_k"] = top_k
if max_tokens != float("inf"):
optional_params["max_output_tokens"] = max_tokens
elif model in model in litellm.vertex_code_text_models:
optional_params["temperature"] = temperature
if max_tokens != float("inf"):
optional_params["max_output_tokens"] = max_tokens
elif custom_llm_provider == "baseten":
optional_params["temperature"] = temperature
optional_params["stream"] = stream
@ -2482,6 +2488,9 @@ class CustomStreamWrapper:
elif self.model in litellm.nlp_cloud_models or self.custom_llm_provider == "nlp_cloud":
chunk = next(self.completion_stream)
completion_obj["content"] = self.handle_nlp_cloud_chunk(chunk)
elif self.model in (litellm.vertex_chat_models + litellm.vertex_code_chat_models + litellm.vertex_text_models + litellm.vertex_code_text_models):
chunk = next(self.completion_stream)
completion_obj["content"] = str(chunk)
elif self.model in litellm.cohere_models or self.custom_llm_provider == "cohere":
chunk = next(self.completion_stream)
completion_obj["content"] = self.handle_cohere_chunk(chunk)