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

(feat) add --cost as a flag to the proxy server cli

Browse source
This commit is contained in:
ishaan-jaff 2023-10-09 15:05:15 -07:00
parent 70720c255e
commit ba754a07a3
2 changed files with 15 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

10
litellm/proxy/proxy_cli.py
View file

@ -74,15 +74,16 @@ def open_config():
@click.option('--config', is_flag=True, help='Create and open .env file from .env.template') @click.option('--config', is_flag=True, help='Create and open .env file from .env.template')
@click.option('--test', flag_value=True, help='proxy chat completions url to make a test request to') @click.option('--test', flag_value=True, help='proxy chat completions url to make a test request to')
@click.option('--local', is_flag=True, default=False, help='for local debugging') @click.option('--local', is_flag=True, default=False, help='for local debugging')
def run_server(host, port, api_base, model, deploy, debug, temperature, max_tokens, drop_params, add_function_to_prompt, max_budget, telemetry, config, test, local): @click.option('--cost', is_flag=True, default=False, help='for viewing cost logs')
def run_server(host, port, api_base, model, deploy, debug, temperature, max_tokens, drop_params, add_function_to_prompt, max_budget, telemetry, config, test, local, cost):
if config: if config:
open_config() open_config()
if local: if local:
from proxy_server import app, initialize, deploy_proxy from proxy_server import app, initialize, deploy_proxy, print_cost_logs
debug = True debug = True
else: else:
from .proxy_server import app, initialize, deploy_proxy from .proxy_server import app, initialize, deploy_proxy, print_cost_logs
if deploy == True: if deploy == True:
print(f"\033[32mLiteLLM: Deploying your proxy to api.litellm.ai\033[0m\n") print(f"\033[32mLiteLLM: Deploying your proxy to api.litellm.ai\033[0m\n")
@ -95,6 +96,9 @@ def run_server(host, port, api_base, model, deploy, debug, temperature, max_toke
return return
if model and "ollama" in model: if model and "ollama" in model:
run_ollama_serve() run_ollama_serve()
if cost == True:
print_cost_logs()
return
if test != False: if test != False:
click.echo('LiteLLM: Making a test ChatCompletions request to your proxy') click.echo('LiteLLM: Making a test ChatCompletions request to your proxy')
import openai import openai

14
litellm/proxy/proxy_server.py
View file

@ -215,16 +215,18 @@ async def track_cost(response):
format='%(asctime)s - %(message)s', format='%(asctime)s - %(message)s',
datefmt='%Y-%m-%d %H:%M:%S' datefmt='%Y-%m-%d %H:%M:%S'
) )
import datetime
response_cost = litellm.completion_cost(completion_response=response) response_cost = litellm.completion_cost(completion_response=response)
logging.info(f"Model {response.model} Cost: ${response_cost:.8f}")
logging.info(f"Model {response.model} Cost: {response_cost:.8f}")
except: except:
pass pass
def print_cost_logs():
with open('cost.log', 'r') as f:
# print this in green
print("\033[1;32m")
print(f.read())
print("\033[0m")
return
@router.get("/ollama_logs") @router.get("/ollama_logs")
async def retrieve_server_log(request: Request): async def retrieve_server_log(request: Request):