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

fix(utils.py): add lru-cache logic to _select_tokenizer

Browse source 
speed up tokenizer load times
This commit is contained in:
Krrish Dholakia 2024-05-14 16:38:45 -07:00
parent 811406a2c6
commit 2b41f09268
1 changed files with 2 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

4
litellm/utils.py
View file

@ -12,10 +12,9 @@ import dotenv, json, traceback, threading, base64, ast
import subprocess, os
from os.path import abspath, join, dirname
import litellm, openai
import itertools
import random, uuid, requests # type: ignore
from functools import wraps
from functools import wraps, lru_cache
import datetime, time
import tiktoken
import uuid
@ -3881,6 +3880,7 @@ def get_replicate_completion_pricing(completion_response=None, total_time=0.0):
return a100_80gb_price_per_second_public * total_time / 1000
@lru_cache(maxsize=128)
def _select_tokenizer(model: str):
from importlib import resources