Merge pull request #1172 from seehi/feat-rag-embedding

Make RAG embedding configurable and add gpt-4-turbo in token_counter.
This commit is contained in:
Alexander Wu 2024-04-10 15:59:31 +08:00 committed by GitHub
commit 63ca5452bc
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
12 changed files with 268 additions and 37 deletions

View file

@ -20,3 +20,18 @@ def run_coroutine_in_new_loop(coroutine) -> Any:
new_loop.call_soon_threadsafe(new_loop.stop)
t.join()
new_loop.close()
class NestAsyncio:
"""Make asyncio event loop reentrant."""
is_applied = False
@classmethod
def apply_once(cls):
"""Ensures `nest_asyncio.apply()` is called only once."""
if not cls.is_applied:
import nest_asyncio
nest_asyncio.apply()
cls.is_applied = True

View file

@ -28,6 +28,7 @@ TOKEN_COSTS = {
"gpt-4-32k-0314": {"prompt": 0.06, "completion": 0.12},
"gpt-4-0613": {"prompt": 0.06, "completion": 0.12},
"gpt-4-turbo-preview": {"prompt": 0.01, "completion": 0.03},
"gpt-4-turbo": {"prompt": 0.01, "completion": 0.03},
"gpt-4-0125-preview": {"prompt": 0.01, "completion": 0.03},
"gpt-4-1106-preview": {"prompt": 0.01, "completion": 0.03},
"gpt-4-vision-preview": {"prompt": 0.01, "completion": 0.03}, # TODO add extra image price calculator
@ -147,6 +148,7 @@ FIREWORKS_GRADE_TOKEN_COSTS = {
TOKEN_MAX = {
"gpt-4-0125-preview": 128000,
"gpt-4-turbo-preview": 128000,
"gpt-4-turbo": 128000,
"gpt-4-1106-preview": 128000,
"gpt-4-vision-preview": 128000,
"gpt-4-1106-vision-preview": 128000,
@ -202,6 +204,7 @@ def count_message_tokens(messages, model="gpt-3.5-turbo-0125"):
"gpt-4-32k-0314",
"gpt-4-0613",
"gpt-4-32k-0613",
"gpt-4-turbo",
"gpt-4-turbo-preview",
"gpt-4-0125-preview",
"gpt-4-1106-preview",