clean up

.env

@@ -5,7 +5,7 @@ LOG_CHAT=false
 
 # `OPENAI_BASE_URL`: 转发openai风格的任何服务地址，允许指定多个, 以逗号隔开。
 # 如果指定超过一个，则任何OPENAI_ROUTE_PREFIX/EXTRA_ROUTE_PREFIX都不能为根路由/
-OPENAI_BASE_URL=https://api.openai-forward.com
+OPENAI_BASE_URL=https://api.openai.com
 
 # `OPENAI_ROUTE_PREFIX`: 可指定所有openai风格(为记录日志)服务的转发路由前缀
 OPENAI_ROUTE_PREFIX=
@@ -18,12 +18,12 @@ EXTRA_BASE_URL=
 # `EXTRA_ROUTE_PREFIX`: 与 EXTRA_BASE_URL 匹配的路由前缀
 EXTRA_ROUTE_PREFIX=
 
-# RATE `LIMIT`: 指定路由的请求速率限制
+# `RATE_LIMIT`: i.e. RPM 对指定路由的请求速率限制
 # format: {route: ratelimit-string}
 # ratelimit-string format [count] [per|/] [n (optional)] [second|minute|hour|day|month|year] :ref:`ratelimit-string`: https://limits.readthedocs.io/en/stable/quickstart.html#rate-limit-string-notation
 RATE_LIMIT='{
 "/healthz": "1000/2minutes",
-"/v1/chat/completions": "5/second",
+"/v1/chat/completions": "300/minute"
 }'
 
 #`RATE_LIMIT_STRATEGY` Options: (fixed-window, fixed-window-elastic-expiry, moving-window) ref: https://limits.readthedocs.io/en/latest/strategies.html
@@ -33,8 +33,8 @@ RATE_LIMIT_STRATEGY=moving-window
 # `GLOBAL_RATE_LIMIT`: 所有`RATE_LIMIT`没有指定的路由. 不填默认无限制
 GLOBAL_RATE_LIMIT=
 
-# same as TPM
-TOKEN_RATE_LIMIT=60/second
+# TPM: 返回的token速率限制
+TOKEN_RATE_LIMIT=50/second
 
 # 设定时区
 TZ=Asia/Shanghai

.env.example

@@ -38,7 +38,7 @@ RATE_LIMIT_STRATEGY=fixed-window
 # `PROXY` http代理
 PROXY=http://localhost:7890
 
-# `TOKEN_RATE_LIMIT` 对每一份流式请求的token速率限制 (注：这里的token并不是gpt中定义的token，而是SSE的chunk)
+# `TOKEN_RATE_LIMIT` 对每一份流式返回的token速率限制 (注：这里的token并不严格等于gpt中定义的token，而是SSE的chunk)
 TOKEN_RATE_LIMIT=16/second
 
 # 设定时区

openai_forward/forwarding/base.py

@@ -264,7 +264,6 @@ async def aiter_bytes(
                 current_time = time.perf_counter()
                 delta = current_time - start_time
                 sleep_time = TOKEN_INTERVAL - delta
-                print(f"{delta=} {sleep_time=}")
                 if sleep_time > 0:
                     time.sleep(sleep_time)
                 start_time = time.perf_counter()

openai_forward/forwarding/settings.py

@@ -3,7 +3,7 @@
 import limits
 from fastapi import Request
 
-from ..config import setting_log
+from ..config import print_startup_info, setting_log, show_rate_limit_info
 from ..helper import env2dict, env2list, format_route_prefix, get_client_ip
 
 ENV_VAR_SEP = ","
@@ -60,7 +60,6 @@ def dynamic_rate_limit(key: str):
 else:
     TOKEN_INTERVAL = 0
 
-from ..config import print_startup_info, show_rate_limit_info
 
 for base_url, route_prefix in zip(OPENAI_BASE_URL, OPENAI_ROUTE_PREFIX):
     print_startup_info(base_url, route_prefix, OPENAI_API_KEY, FWD_KEY, LOG_CHAT)

pytest.ini

@@ -5,4 +5,4 @@ markers =
     timeout: marks test timeout duration
     repeat: marks that test run n times
 addopts = --doctest-modules --doctest-glob=README.md --doctest-glob=*.py --ignore=setup.py
-norecursedirs = Examples ssl
+norecursedirs = Examples ssl scripts