Revert changes to large text

Author: tomukmatthews

tests/conftest.py

@@ -23,6 +23,7 @@
 from litellm import create_pretrained_tokenizer, decode, encode, get_modified_max_tokens
 from litellm import token_counter as token_counter_old
 from litellm.litellm_core_utils.token_counter import token_counter as token_counter_new
+from tests.large_text import text
 
 
 def token_counter_both_assert_same(**args):
@@ -276,21 +277,22 @@ def test_gpt_vision_token_counting():
 @pytest.mark.parametrize(
     "model",
     [
+        "gpt-4-vision-preview",
         "gpt-4o",
         "claude-3-opus-20240229",
         "command-nightly",
         "mistral/mistral-tiny",
     ],
 )
-def test_load_test_token_counter(model, large_text):
+def test_load_test_token_counter(model):
     """
     Token count large prompt 100 times.
 
     Assert time taken is < 1.5s.
     """
     import tiktoken
 
-    messages = [{"role": "user", "content": large_text}] * 10
+    messages = [{"role": "user", "content": text}] * 10
 
     start_time = time.time()
     for _ in range(10):

tests/large_text.py

@@ -683,15 +683,15 @@ async def test2():
 
 @pytest.mark.asyncio
 @pytest.mark.parametrize("sync_mode", [True, False])
-async def test_async_router_context_window_fallback(sync_mode, large_text):
+async def test_async_router_context_window_fallback(sync_mode):
     """
     - Give a gpt-4 model group with different context windows (8192k vs. 128k)
     - Send a 10k prompt
     - Assert it works
     """
     import os
 
-    text = large_text
+    from large_text import text
 
     litellm.set_verbose = False
     litellm._turn_on_debug()
@@ -775,17 +775,15 @@ def test_router_rpm_pre_call_check():
         pytest.fail(f"Got unexpected exception on router! - {str(e)}")
 
 
-def test_router_context_window_check_pre_call_check_in_group_custom_model_info(
-    large_text,
-):
+def test_router_context_window_check_pre_call_check_in_group_custom_model_info():
     """
     - Give a gpt-3.5-turbo model group with different context windows (4k vs. 16k)
     - Send a 5k prompt
     - Assert it works
     """
     import os
 
-    text = large_text
+    from large_text import text
 
     litellm.set_verbose = False
 
@@ -831,15 +829,15 @@ def test_router_context_window_check_pre_call_check_in_group_custom_model_info(
         pytest.fail(f"Got unexpected exception on router! - {str(e)}")
 
 
-def test_router_context_window_check_pre_call_check(large_text):
+def test_router_context_window_check_pre_call_check():
     """
     - Give a gpt-3.5-turbo model group with different context windows (4k vs. 16k)
     - Send a 5k prompt
     - Assert it works
     """
     import os
 
-    text = large_text
+    from large_text import text
 
     litellm.set_verbose = False
 
@@ -885,15 +883,15 @@ def test_router_context_window_check_pre_call_check(large_text):
         pytest.fail(f"Got unexpected exception on router! - {str(e)}")
 
 
-def test_router_context_window_check_pre_call_check_out_group(large_text):
+def test_router_context_window_check_pre_call_check_out_group():
     """
     - Give 2 gpt-3.5-turbo model groups with different context windows (4k vs. 16k)
     - Send a 5k prompt
     - Assert it works
     """
     import os
 
-    text = large_text
+    from large_text import text
 
     litellm.set_verbose = False
 
@@ -1138,7 +1136,7 @@ async def test_aimg_gen_on_router():
                     "api_base": os.getenv("AZURE_SWEDEN_API_BASE"),
                     "api_key": os.getenv("AZURE_SWEDEN_API_KEY"),
                 },
-            },
+            }
         ]
         router = Router(model_list=model_list, num_retries=3)
         response = await router.aimage_generation(
@@ -2789,5 +2787,4 @@ def test_router_get_model_group_info():
     assert model_group_info is not None
     assert model_group_info.model_group == "gpt-4"
     assert model_group_info.input_cost_per_token > 0
-
-    assert model_group_info.output_cost_per_token > 0
+    assert model_group_info.output_cost_per_token > 0

tests/litellm/litellm_core_utils/test_token_counter.py

@@ -4,6 +4,7 @@
 from openai import AsyncOpenAI, AsyncAzureOpenAI
 import uuid
 import traceback
+from large_text import text
 from dotenv import load_dotenv
 from statistics import mean, median
 

tests/local_testing/test_router.py

@@ -3,6 +3,7 @@
 import pytest
 import asyncio
 import aiohttp
+from large_text import text
 import time
 from typing import Optional