phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 03:04:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge pull request #4925 from BerriAI/litellm_vertex_mistral

Browse source 
feat(vertex_ai_partner.py): Vertex AI Mistral Support
This commit is contained in:
Krish Dholakia 2024-07-27 21:51:26 -07:00 • committed by GitHub
commit e3a94ac013
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
10 changed files with 365 additions and 147 deletions
Download patch file Download diff file Expand all files Collapse all files

9
litellm/main.py
View file

@ -121,7 +121,7 @@ from .llms.prompt_templates.factory import (
)
from .llms.text_completion_codestral import CodestralTextCompletion
from .llms.triton import TritonChatCompletion
from .llms.vertex_ai_llama import VertexAILlama3
from .llms.vertex_ai_partner import VertexAIPartnerModels
from .llms.vertex_httpx import VertexLLM
from .llms.watsonx import IBMWatsonXAI
from .types.llms.openai import HttpxBinaryResponseContent
@ -158,7 +158,7 @@ triton_chat_completions = TritonChatCompletion()
bedrock_chat_completion = BedrockLLM()
bedrock_converse_chat_completion = BedrockConverseLLM()
vertex_chat_completion = VertexLLM()
vertex_llama_chat_completion = VertexAILlama3()
vertex_partner_models_chat_completion = VertexAIPartnerModels()
watsonxai = IBMWatsonXAI()
####### COMPLETION ENDPOINTS ################
@ -1867,6 +1867,7 @@ def completion(
custom_prompt_dict=custom_prompt_dict,
client=client, # pass AsyncOpenAI, OpenAI client
encoding=encoding,
custom_llm_provider="databricks",
)
except Exception as e:
## LOGGING - log the original exception returned
@ -2068,8 +2069,8 @@ def completion(
timeout=timeout,
client=client,
)
elif model.startswith("meta/"):
model_response = vertex_llama_chat_completion.completion(
elif model.startswith("meta/") or model.startswith("mistral"):
model_response = vertex_partner_models_chat_completion.completion(
model=model,
messages=messages,
model_response=model_response,