phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 11:14:04 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Revert "changes"

Browse source 
This reverts commit fa88bc9632.
This commit is contained in:
Utkash Dubey 2025-03-10 14:46:21 -07:00
parent 1f17daf52c
commit c7a04140e9
19 changed files with 141 additions and 191 deletions
Download patch file Download diff file Expand all files Collapse all files

70
tests/local_testing/test_completion_cost.py
View file

@ -634,7 +634,8 @@ def test_gemini_completion_cost(above_128k, provider):
"""
Check if cost correctly calculated for gemini models based on context window
"""
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
if provider == "gemini":
model_name = "gemini-1.5-flash-latest"
else:
@ -689,7 +690,8 @@ def _count_characters(text):
def test_vertex_ai_completion_cost():
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
text = "The quick brown fox jumps over the lazy dog."
characters = _count_characters(text=text)
@ -724,7 +726,8 @@ def test_vertex_ai_medlm_completion_cost():
model=model, messages=messages, custom_llm_provider="vertex_ai"
)
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
model = "vertex_ai/medlm-medium"
messages = [{"role": "user", "content": "Test MedLM completion cost."}]
@ -743,7 +746,8 @@ def test_vertex_ai_claude_completion_cost():
from litellm import Choices, Message, ModelResponse
from litellm.utils import Usage
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
litellm.set_verbose = True
input_tokens = litellm.token_counter(
@ -792,7 +796,8 @@ def test_vertex_ai_embedding_completion_cost(caplog):
"""
Relevant issue - https://github.com/BerriAI/litellm/issues/4630
"""
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
text = "The quick brown fox jumps over the lazy dog."
input_tokens = litellm.token_counter(
@ -834,7 +839,8 @@ def test_vertex_ai_embedding_completion_cost(caplog):
# from test_amazing_vertex_completion import load_vertex_ai_credentials
# load_vertex_ai_credentials()
# litellm.model_cost = litellm.get_locally_cached_model_cost_map()
# os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
# litellm.model_cost = litellm.get_model_cost_map(url="")
# text = "The quick brown fox jumps over the lazy dog."
# input_tokens = litellm.token_counter(
@ -861,7 +867,8 @@ def test_vertex_ai_embedding_completion_cost(caplog):
def test_completion_azure_ai():
try:
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
litellm.set_verbose = True
response = litellm.completion(
@ -967,7 +974,8 @@ def test_vertex_ai_mistral_predict_cost(usage):
@pytest.mark.parametrize("model", ["openai/tts-1", "azure/tts-1"])
def test_completion_cost_tts(model):
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
cost = completion_cost(
model=model,
@ -1163,7 +1171,8 @@ def test_completion_cost_azure_common_deployment_name():
],
)
def test_completion_cost_prompt_caching(model, custom_llm_provider):
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
from litellm.utils import Choices, Message, ModelResponse, Usage
@ -1264,7 +1273,8 @@ def test_completion_cost_prompt_caching(model, custom_llm_provider):
],
)
def test_completion_cost_databricks(model):
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
model, messages = model, [{"role": "user", "content": "What is 2+2?"}]
resp = litellm.completion(model=model, messages=messages) # works fine
@ -1281,7 +1291,8 @@ def test_completion_cost_databricks(model):
],
)
def test_completion_cost_databricks_embedding(model):
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
resp = litellm.embedding(model=model, input=["hey, how's it going?"]) # works fine
print(resp)
@ -1308,7 +1319,8 @@ def test_get_model_params_fireworks_ai(model, base_model):
["fireworks_ai/llama-v3p1-405b-instruct", "fireworks_ai/mixtral-8x7b-instruct"],
)
def test_completion_cost_fireworks_ai(model):
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
messages = [{"role": "user", "content": "Hey, how's it going?"}]
resp = litellm.completion(model=model, messages=messages) # works fine
@ -1325,7 +1337,8 @@ def test_cost_azure_openai_prompt_caching():
)
from litellm import get_model_info
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
model = "azure/o1-mini"
@ -1416,7 +1429,8 @@ def test_cost_azure_openai_prompt_caching():
def test_completion_cost_vertex_llama3():
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
from litellm.utils import Choices, Message, ModelResponse, Usage
@ -1456,7 +1470,8 @@ def test_cost_openai_prompt_caching():
from litellm.utils import Choices, Message, ModelResponse, Usage
from litellm import get_model_info
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
model = "gpt-4o-mini-2024-07-18"
@ -1546,7 +1561,8 @@ def test_cost_openai_prompt_caching():
def test_completion_cost_azure_ai_rerank(model):
from litellm import RerankResponse, rerank
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
response = RerankResponse(
id="b01dbf2e-63c8-4981-9e69-32241da559ed",
@ -1577,7 +1593,8 @@ def test_completion_cost_azure_ai_rerank(model):
def test_together_ai_embedding_completion_cost():
from litellm.utils import Choices, EmbeddingResponse, Message, ModelResponse, Usage
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
response = EmbeddingResponse(
model="togethercomputer/m2-bert-80M-8k-retrieval",
data=[
@ -2434,7 +2451,8 @@ def test_completion_cost_params_gemini_3():
from litellm.llms.vertex_ai.cost_calculator import cost_per_character
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
response = ModelResponse(
id="chatcmpl-61043504-4439-48be-9996-e29bdee24dc3",
@ -2503,7 +2521,8 @@ def test_completion_cost_params_gemini_3():
# @pytest.mark.flaky(retries=3, delay=1)
@pytest.mark.parametrize("stream", [False]) # True,
async def test_test_completion_cost_gpt4o_audio_output_from_model(stream):
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
from litellm.types.utils import (
Choices,
Message,
@ -2600,7 +2619,8 @@ def test_completion_cost_model_response_cost(response_model, custom_llm_provider
"""
from litellm import ModelResponse
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
litellm.set_verbose = True
response = {
@ -2700,7 +2720,8 @@ def test_select_model_name_for_cost_calc():
def test_moderations():
from litellm import moderation
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
litellm.add_known_models()
assert "omni-moderation-latest" in litellm.model_cost
@ -2753,7 +2774,8 @@ def test_bedrock_cost_calc_with_region():
from litellm import ModelResponse
litellm.model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
litellm.model_cost = litellm.get_model_cost_map(url="")
litellm.add_known_models()
@ -2952,7 +2974,9 @@ async def test_cost_calculator_with_custom_pricing_router(model_item, custom_pri
def test_json_valid_model_cost_map():
import json
model_cost = litellm.get_locally_cached_model_cost_map()
os.environ["LITELLM_LOCAL_MODEL_COST_MAP"] = "True"
model_cost = litellm.get_model_cost_map(url="")
try:
# Attempt to serialize and deserialize the JSON