Spaces:
Sleeping
Sleeping
File size: 8,612 Bytes
7db0ae4 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 |
#### What this tests ####
# This tests the the acompletion function #
import sys, os
import pytest
import traceback
import asyncio, logging
sys.path.insert(
0, os.path.abspath("../..")
) # Adds the parent directory to the system path
import litellm
from litellm import completion, acompletion, acreate
litellm.num_retries = 3
def test_sync_response_anyscale():
litellm.set_verbose = False
user_message = "Hello, how are you?"
messages = [{"content": user_message, "role": "user"}]
try:
response = completion(
model="anyscale/mistralai/Mistral-7B-Instruct-v0.1",
messages=messages,
timeout=5,
)
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"An exception occurred: {e}")
# test_sync_response_anyscale()
def test_async_response_openai():
import asyncio
litellm.set_verbose = True
async def test_get_response():
user_message = "Hello, how are you?"
messages = [{"content": user_message, "role": "user"}]
try:
response = await acompletion(
model="gpt-3.5-turbo", messages=messages, timeout=5
)
print(f"response: {response}")
print(f"response ms: {response._response_ms}")
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"An exception occurred: {e}")
print(e)
asyncio.run(test_get_response())
# test_async_response_openai()
def test_async_response_azure():
import asyncio
litellm.set_verbose = True
async def test_get_response():
user_message = "What do you know?"
messages = [{"content": user_message, "role": "user"}]
try:
response = await acompletion(
model="azure/gpt-turbo",
messages=messages,
base_url=os.getenv("CLOUDFLARE_AZURE_BASE_URL"),
api_key=os.getenv("AZURE_FRANCE_API_KEY"),
)
print(f"response: {response}")
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"An exception occurred: {e}")
asyncio.run(test_get_response())
# test_async_response_azure()
def test_async_anyscale_response():
import asyncio
litellm.set_verbose = True
async def test_get_response():
user_message = "Hello, how are you?"
messages = [{"content": user_message, "role": "user"}]
try:
response = await acompletion(
model="anyscale/mistralai/Mistral-7B-Instruct-v0.1",
messages=messages,
timeout=5,
)
# response = await response
print(f"response: {response}")
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"An exception occurred: {e}")
asyncio.run(test_get_response())
# test_async_anyscale_response()
@pytest.mark.skip(reason="Flaky test-cloudflare is very unstable")
def test_async_completion_cloudflare():
try:
litellm.set_verbose = True
async def test():
response = await litellm.acompletion(
model="cloudflare/@cf/meta/llama-2-7b-chat-int8",
messages=[{"content": "what llm are you", "role": "user"}],
max_tokens=5,
num_retries=3,
)
print(response)
return response
response = asyncio.run(test())
text_response = response["choices"][0]["message"]["content"]
assert len(text_response) > 1 # more than 1 chars in response
except Exception as e:
pytest.fail(f"Error occurred: {e}")
# test_async_completion_cloudflare()
@pytest.mark.skip(reason="Flaky test")
def test_get_cloudflare_response_streaming():
import asyncio
async def test_async_call():
user_message = "write a short poem in one sentence"
messages = [{"content": user_message, "role": "user"}]
try:
litellm.set_verbose = False
response = await acompletion(
model="cloudflare/@cf/meta/llama-2-7b-chat-int8",
messages=messages,
stream=True,
num_retries=3, # cloudflare ai workers is EXTREMELY UNSTABLE
)
print(type(response))
import inspect
is_async_generator = inspect.isasyncgen(response)
print(is_async_generator)
output = ""
i = 0
async for chunk in response:
print(chunk)
token = chunk["choices"][0]["delta"].get("content", "")
if token == None:
continue # openai v1.0.0 returns content=None
output += token
assert output is not None, "output cannot be None."
assert isinstance(output, str), "output needs to be of type str"
assert len(output) > 0, "Length of output needs to be greater than 0."
print(f"output: {output}")
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"An exception occurred: {e}")
asyncio.run(test_async_call())
@pytest.mark.asyncio
async def test_hf_completion_tgi():
# litellm.set_verbose=True
try:
response = await acompletion(
model="huggingface/HuggingFaceH4/zephyr-7b-beta",
messages=[{"content": "Hello, how are you?", "role": "user"}],
)
# Add any assertions here to check the response
print(response)
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"Error occurred: {e}")
# test_get_cloudflare_response_streaming()
def test_get_response_streaming():
import asyncio
async def test_async_call():
user_message = "write a short poem in one sentence"
messages = [{"content": user_message, "role": "user"}]
try:
litellm.set_verbose = True
response = await acompletion(
model="gpt-3.5-turbo", messages=messages, stream=True, timeout=5
)
print(type(response))
import inspect
is_async_generator = inspect.isasyncgen(response)
print(is_async_generator)
output = ""
i = 0
async for chunk in response:
token = chunk["choices"][0]["delta"].get("content", "")
if token == None:
continue # openai v1.0.0 returns content=None
output += token
assert output is not None, "output cannot be None."
assert isinstance(output, str), "output needs to be of type str"
assert len(output) > 0, "Length of output needs to be greater than 0."
print(f"output: {output}")
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"An exception occurred: {e}")
asyncio.run(test_async_call())
# test_get_response_streaming()
def test_get_response_non_openai_streaming():
import asyncio
litellm.set_verbose = True
litellm.num_retries = 0
async def test_async_call():
user_message = "Hello, how are you?"
messages = [{"content": user_message, "role": "user"}]
try:
response = await acompletion(
model="anyscale/mistralai/Mistral-7B-Instruct-v0.1",
messages=messages,
stream=True,
timeout=5,
)
print(type(response))
import inspect
is_async_generator = inspect.isasyncgen(response)
print(is_async_generator)
output = ""
i = 0
async for chunk in response:
token = chunk["choices"][0]["delta"].get("content", None)
if token == None:
continue
print(token)
output += token
print(f"output: {output}")
assert output is not None, "output cannot be None."
assert isinstance(output, str), "output needs to be of type str"
assert len(output) > 0, "Length of output needs to be greater than 0."
except litellm.Timeout as e:
pass
except Exception as e:
pytest.fail(f"An exception occurred: {e}")
return response
asyncio.run(test_async_call())
# test_get_response_non_openai_streaming()
|