phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

Added openrouter.ai support

Browse source
This commit is contained in:
Zakhar Kogan 2023-07-28 11:44:08 +03:00
parent 2463dc234b
commit 00c2384a2d
8 changed files with 118 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

5
.env.example Normal file
View file

@ -0,0 +1,5 @@
OPENAI_API_KEY = ""
COHERE_API_KEY = ""
OPENROUTER_API_KEY = ""
OR_SITE_URL = ""
OR_APP_NAME = "LiteLLM Example app"

1
.gitignore vendored Normal file
View file

@ -0,0 +1 @@
.env

60
build/lib/litellm/main.py
View file

@ -1,4 +1,7 @@
import os, openai, cohere import os, openai, cohere, dotenv
# Loading env variables using dotenv
dotenv.load_dotenv()
####### COMPLETION MODELS ################### ####### COMPLETION MODELS ###################
open_ai_chat_completion_models = [ open_ai_chat_completion_models = [
@ -13,6 +16,18 @@ cohere_models = [
'command-nightly', 'command-nightly',
] ]
openrouter_models = [
'google/palm-2-codechat-bison',
'google/palm-2-chat-bison',
'openai/gpt-3.5-turbo',
'openai/gpt-3.5-turbo-16k',
'openai/gpt-4-32k',
'anthropic/claude-2',
'anthropic/claude-instant-v1',
'meta-llama/llama-2-13b-chat',
'meta-llama/llama-2-70b-chat'
]
####### EMBEDDING MODELS ################### ####### EMBEDDING MODELS ###################
open_ai_embedding_models = [ open_ai_embedding_models = [
'text-embedding-ada-002' 'text-embedding-ada-002'
@ -34,6 +49,32 @@ def completion(model, messages, azure=False):
engine=model, engine=model,
messages = messages messages = messages
) )
elif "replicate" in model:
prompt = " ".join([message["content"] for message in messages])
output = replicate.run(
model,
input={
"prompt": prompt,
})
print(f"output: {output}")
response = ""
for item in output:
print(f"item: {item}")
response += item
new_response = {
"choices": [
{
"finish_reason": "stop",
"index": 0,
"message": {
"content": response,
"role": "assistant"
}
}
]
}
print(f"new response: {new_response}")
response = new_response
elif model in cohere_models: elif model in cohere_models:
cohere_key = os.environ.get("COHERE_API_KEY") cohere_key = os.environ.get("COHERE_API_KEY")
co = cohere.Client(cohere_key) co = cohere.Client(cohere_key)
@ -76,6 +117,22 @@ def completion(model, messages, azure=False):
model=model, model=model,
prompt = prompt prompt = prompt
) )
elif model in openrouter_models:
openai.api_base = "https://openrouter.ai/api/v1"
openai.api_key = os.environ.get("OPENROUTER_API_KEY")
prompt = " ".join([message["content"] for message in messages])
response = openai.ChatCompletion.create(
model=model,
messages=messages,
headers={
"HTTP-Referer": os.environ.get("OR_SITE_URL"), # To identify your app
"X-Title": os.environ.get("OR_APP_NAME")
},
)
reply = response.choices[0].message
return response return response
@ -100,4 +157,3 @@ def embedding(model, input=[], azure=False):
############################################# #############################################
############################################# #############################################

38
completion_test.py
View file

@ -1,4 +1,4 @@
from main import completion from litellm.main import completion
import os import os
## Configs for Models ## ## Configs for Models ##
@ -8,23 +8,27 @@ import os
messages = [{ "content": "Hello, how are you?","role": "user"}] messages = [{ "content": "Hello, how are you?","role": "user"}]
# openai call # # openai call
response = completion(model="gpt-3.5-turbo", messages=messages) # response = completion(model="gpt-3.5-turbo", messages=messages)
print("\nOpenAI call") # print("\nOpenAI call")
print(response) # print(response)
# azure openai call # # azure openai call
response = completion("chatgpt-test", messages, azure=True) # response = completion("chatgpt-test", messages, azure=True)
print("\nAzure call") # print("\nAzure call")
print(response) # print(response)
# text davinci openai call # # text davinci openai call
response = completion("text-davinci-003", messages) # response = completion("text-davinci-003", messages)
print("\nDavinci call") # print("\nDavinci call")
print(response) # print(response)
# cohere call # # cohere call
response = completion("command-nightly", messages) # response = completion("command-nightly", messages)
print("\nCohere call") # print("\nCohere call")
print(response) # print(response)
# openrouter call
response = completion("google/palm-2-codechat-bison", messages)
print("\OpenRouter call")
print(response)

BIN
litellm/__pycache__/__init__.cpython-311.pyc Normal file
View file

Binary file not shown.

BIN
litellm/__pycache__/main.cpython-311.pyc Normal file
View file

Binary file not shown.

33
litellm/main.py
View file

@ -1,4 +1,7 @@
import os, openai, cohere import os, openai, cohere, dotenv
# Loading env variables using dotenv
dotenv.load_dotenv()
####### COMPLETION MODELS ################### ####### COMPLETION MODELS ###################
open_ai_chat_completion_models = [ open_ai_chat_completion_models = [
@ -13,6 +16,18 @@ cohere_models = [
'command-nightly', 'command-nightly',
] ]
openrouter_models = [
'google/palm-2-codechat-bison',
'google/palm-2-chat-bison',
'openai/gpt-3.5-turbo',
'openai/gpt-3.5-turbo-16k',
'openai/gpt-4-32k',
'anthropic/claude-2',
'anthropic/claude-instant-v1',
'meta-llama/llama-2-13b-chat',
'meta-llama/llama-2-70b-chat'
]
####### EMBEDDING MODELS ################### ####### EMBEDDING MODELS ###################
open_ai_embedding_models = [ open_ai_embedding_models = [
'text-embedding-ada-002' 'text-embedding-ada-002'
@ -102,6 +117,22 @@ def completion(model, messages, azure=False):
model=model, model=model,
prompt = prompt prompt = prompt
) )
elif model in openrouter_models:
openai.api_base = "https://openrouter.ai/api/v1"
openai.api_key = os.environ.get("OPENROUTER_API_KEY")
prompt = " ".join([message["content"] for message in messages])
response = openai.ChatCompletion.create(
model=model,
messages=messages,
headers={
"HTTP-Referer": os.environ.get("OR_SITE_URL"), # To identify your app
"X-Title": os.environ.get("OR_APP_NAME")
},
)
reply = response.choices[0].message
return response return response

0
testos.py Normal file
View file