mirror of
https://github.com/BerriAI/litellm.git
synced 2026-10-08 03:08:45 +00:00
rename proxy track cost callback test
This commit is contained in:
Ishaan Jaff
parent
bbdb9c3b57
commit
ec7f0ce2d0
2 changed files with 39 additions and 41 deletions
|
|
@ -36,7 +36,7 @@ import TabItem from '@theme/TabItem';
|
|||
- Virtual Key Rate Limit
|
||||
- User Rate Limit
|
||||
- Team Limit
|
||||
- The `_PROXY_track_cost_callback` updates spend / usage in the LiteLLM database. [Here is everything tracked in the DB per request](https://github.com/BerriAI/litellm/blob/ba41a72f92a9abf1d659a87ec880e8e319f87481/schema.prisma#L172)
|
||||
- The `_ProxyDBLogger` updates spend / usage in the LiteLLM database. [Here is everything tracked in the DB per request](https://github.com/BerriAI/litellm/blob/ba41a72f92a9abf1d659a87ec880e8e319f87481/schema.prisma#L172)
|
||||
|
||||
## Frequently Asked Questions
|
||||
|
||||
|
|
|
|||
|
|
@ -507,9 +507,9 @@ def test_call_with_user_over_budget(prisma_client):
|
|||
|
||||
# update spend using track_cost callback, make 2nd request, it should fail
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
resp = ModelResponse(
|
||||
id="chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac",
|
||||
|
|
@ -526,7 +526,7 @@ def test_call_with_user_over_budget(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"stream": False,
|
||||
"litellm_params": {
|
||||
|
|
@ -604,9 +604,9 @@ def test_call_with_end_user_over_budget(prisma_client):
|
|||
|
||||
# update spend using track_cost callback, make 2nd request, it should fail
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
resp = ModelResponse(
|
||||
id="chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac",
|
||||
|
|
@ -623,7 +623,7 @@ def test_call_with_end_user_over_budget(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"stream": False,
|
||||
"litellm_params": {
|
||||
|
|
@ -711,9 +711,9 @@ def test_call_with_proxy_over_budget(prisma_client):
|
|||
|
||||
# update spend using track_cost callback, make 2nd request, it should fail
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
resp = ModelResponse(
|
||||
id="chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac",
|
||||
|
|
@ -730,7 +730,7 @@ def test_call_with_proxy_over_budget(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"stream": False,
|
||||
"litellm_params": {
|
||||
|
|
@ -802,9 +802,9 @@ def test_call_with_user_over_budget_stream(prisma_client):
|
|||
|
||||
# update spend using track_cost callback, make 2nd request, it should fail
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
resp = ModelResponse(
|
||||
id="chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac",
|
||||
|
|
@ -821,7 +821,7 @@ def test_call_with_user_over_budget_stream(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"stream": True,
|
||||
"complete_streaming_response": resp,
|
||||
|
|
@ -908,9 +908,9 @@ def test_call_with_proxy_over_budget_stream(prisma_client):
|
|||
|
||||
# update spend using track_cost callback, make 2nd request, it should fail
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
resp = ModelResponse(
|
||||
id="chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac",
|
||||
|
|
@ -927,7 +927,7 @@ def test_call_with_proxy_over_budget_stream(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"stream": True,
|
||||
"complete_streaming_response": resp,
|
||||
|
|
@ -1519,9 +1519,9 @@ def test_call_with_key_over_budget(prisma_client):
|
|||
# update spend using track_cost callback, make 2nd request, it should fail
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.caching.caching import Cache
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
litellm.cache = Cache()
|
||||
import time
|
||||
|
|
@ -1544,7 +1544,7 @@ def test_call_with_key_over_budget(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"model": "chatgpt-v-2",
|
||||
"stream": False,
|
||||
|
|
@ -1636,9 +1636,7 @@ def test_call_with_key_over_budget_no_cache(prisma_client):
|
|||
print("result from user auth with new key", result)
|
||||
|
||||
# update spend using track_cost callback, make 2nd request, it should fail
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
from litellm.proxy.proxy_server import user_api_key_cache
|
||||
|
||||
user_api_key_cache.in_memory_cache.cache_dict = {}
|
||||
|
|
@ -1668,7 +1666,8 @@ def test_call_with_key_over_budget_no_cache(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"model": "chatgpt-v-2",
|
||||
"stream": False,
|
||||
|
|
@ -1874,9 +1873,9 @@ async def test_call_with_key_never_over_budget(prisma_client):
|
|||
import uuid
|
||||
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
request_id = f"chatcmpl-{uuid.uuid4()}"
|
||||
|
||||
|
|
@ -1897,7 +1896,7 @@ async def test_call_with_key_never_over_budget(prisma_client):
|
|||
prompt_tokens=210000, completion_tokens=200000, total_tokens=41000
|
||||
),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"model": "chatgpt-v-2",
|
||||
"stream": False,
|
||||
|
|
@ -1965,9 +1964,9 @@ async def test_call_with_key_over_budget_stream(prisma_client):
|
|||
import uuid
|
||||
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
|
||||
request_id = f"chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac{uuid.uuid4()}"
|
||||
resp = ModelResponse(
|
||||
|
|
@ -1985,7 +1984,7 @@ async def test_call_with_key_over_budget_stream(prisma_client):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"call_type": "acompletion",
|
||||
"model": "sagemaker-chatgpt-v-2",
|
||||
|
|
@ -2409,9 +2408,7 @@ async def track_cost_callback_helper_fn(generated_key: str, user_id: str):
|
|||
import uuid
|
||||
|
||||
from litellm import Choices, Message, ModelResponse, Usage
|
||||
from litellm.proxy.proxy_server import (
|
||||
_PROXY_track_cost_callback as track_cost_callback,
|
||||
)
|
||||
from litellm.proxy.proxy_server import _ProxyDBLogger
|
||||
|
||||
request_id = f"chatcmpl-e41836bb-bb8b-4df2-8e70-8f3e160155ac{uuid.uuid4()}"
|
||||
resp = ModelResponse(
|
||||
|
|
@ -2429,7 +2426,8 @@ async def track_cost_callback_helper_fn(generated_key: str, user_id: str):
|
|||
model="gpt-35-turbo", # azure always has model written like this
|
||||
usage=Usage(prompt_tokens=210, completion_tokens=200, total_tokens=410),
|
||||
)
|
||||
await track_cost_callback(
|
||||
proxy_db_logger = _ProxyDBLogger()
|
||||
await proxy_db_logger._PROXY_track_cost_callback(
|
||||
kwargs={
|
||||
"call_type": "acompletion",
|
||||
"model": "sagemaker-chatgpt-v-2",
|
||||
|
|
|
|||
Loading…
Add table
Reference in a new issue