File size: 12,811 Bytes
a8b3f00 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 |
import os
from collections.abc import Generator
import pytest
from core.model_runtime.entities.llm_entities import LLMResult, LLMResultChunk, LLMResultChunkDelta
from core.model_runtime.entities.message_entities import (
AssistantPromptMessage,
PromptMessageTool,
SystemPromptMessage,
UserPromptMessage,
)
from core.model_runtime.errors.validate import CredentialsValidateFailedError
from core.model_runtime.model_providers.xinference.llm.llm import XinferenceAILargeLanguageModel
"""FOR MOCK FIXTURES, DO NOT REMOVE"""
from tests.integration_tests.model_runtime.__mock.openai import setup_openai_mock
from tests.integration_tests.model_runtime.__mock.xinference import setup_xinference_mock
@pytest.mark.parametrize(("setup_openai_mock", "setup_xinference_mock"), [("chat", "none")], indirect=True)
def test_validate_credentials_for_chat_model(setup_openai_mock, setup_xinference_mock):
model = XinferenceAILargeLanguageModel()
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(
model="ChatGLM3",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": "www " + os.environ.get("XINFERENCE_CHAT_MODEL_UID"),
},
)
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(model="aaaaa", credentials={"server_url": "", "model_uid": ""})
model.validate_credentials(
model="ChatGLM3",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_CHAT_MODEL_UID"),
},
)
@pytest.mark.parametrize(("setup_openai_mock", "setup_xinference_mock"), [("chat", "none")], indirect=True)
def test_invoke_chat_model(setup_openai_mock, setup_xinference_mock):
model = XinferenceAILargeLanguageModel()
response = model.invoke(
model="ChatGLM3",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_CHAT_MODEL_UID"),
},
prompt_messages=[
SystemPromptMessage(
content="You are a helpful AI assistant.",
),
UserPromptMessage(content="Hello World!"),
],
model_parameters={
"temperature": 0.7,
"top_p": 1.0,
},
stop=["you"],
user="abc-123",
stream=False,
)
assert isinstance(response, LLMResult)
assert len(response.message.content) > 0
assert response.usage.total_tokens > 0
@pytest.mark.parametrize(("setup_openai_mock", "setup_xinference_mock"), [("chat", "none")], indirect=True)
def test_invoke_stream_chat_model(setup_openai_mock, setup_xinference_mock):
model = XinferenceAILargeLanguageModel()
response = model.invoke(
model="ChatGLM3",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_CHAT_MODEL_UID"),
},
prompt_messages=[
SystemPromptMessage(
content="You are a helpful AI assistant.",
),
UserPromptMessage(content="Hello World!"),
],
model_parameters={
"temperature": 0.7,
"top_p": 1.0,
},
stop=["you"],
stream=True,
user="abc-123",
)
assert isinstance(response, Generator)
for chunk in response:
assert isinstance(chunk, LLMResultChunk)
assert isinstance(chunk.delta, LLMResultChunkDelta)
assert isinstance(chunk.delta.message, AssistantPromptMessage)
assert len(chunk.delta.message.content) > 0 if chunk.delta.finish_reason is None else True
"""
Function calling of xinference does not support stream mode currently
"""
# def test_invoke_stream_chat_model_with_functions():
# model = XinferenceAILargeLanguageModel()
# response = model.invoke(
# model='ChatGLM3-6b',
# credentials={
# 'server_url': os.environ.get('XINFERENCE_SERVER_URL'),
# 'model_type': 'text-generation',
# 'model_name': 'ChatGLM3',
# 'model_uid': os.environ.get('XINFERENCE_CHAT_MODEL_UID')
# },
# prompt_messages=[
# SystemPromptMessage(
# content='你是一个天气机器人，可以通过调用函数来获取天气信息',
# ),
# UserPromptMessage(
# content='波士顿天气如何？'
# )
# ],
# model_parameters={
# 'temperature': 0,
# 'top_p': 1.0,
# },
# stop=['you'],
# user='abc-123',
# stream=True,
# tools=[
# PromptMessageTool(
# name='get_current_weather',
# description='Get the current weather in a given location',
# parameters={
# "type": "object",
# "properties": {
# "location": {
# "type": "string",
# "description": "The city and state e.g. San Francisco, CA"
# },
# "unit": {
# "type": "string",
# "enum": ["celsius", "fahrenheit"]
# }
# },
# "required": [
# "location"
# ]
# }
# )
# ]
# )
# assert isinstance(response, Generator)
# call: LLMResultChunk = None
# chunks = []
# for chunk in response:
# chunks.append(chunk)
# assert isinstance(chunk, LLMResultChunk)
# assert isinstance(chunk.delta, LLMResultChunkDelta)
# assert isinstance(chunk.delta.message, AssistantPromptMessage)
# assert len(chunk.delta.message.content) > 0 if chunk.delta.finish_reason is None else True
# if chunk.delta.message.tool_calls and len(chunk.delta.message.tool_calls) > 0:
# call = chunk
# break
# assert call is not None
# assert call.delta.message.tool_calls[0].function.name == 'get_current_weather'
# def test_invoke_chat_model_with_functions():
# model = XinferenceAILargeLanguageModel()
# response = model.invoke(
# model='ChatGLM3-6b',
# credentials={
# 'server_url': os.environ.get('XINFERENCE_SERVER_URL'),
# 'model_type': 'text-generation',
# 'model_name': 'ChatGLM3',
# 'model_uid': os.environ.get('XINFERENCE_CHAT_MODEL_UID')
# },
# prompt_messages=[
# UserPromptMessage(
# content='What is the weather like in San Francisco?'
# )
# ],
# model_parameters={
# 'temperature': 0.7,
# 'top_p': 1.0,
# },
# stop=['you'],
# user='abc-123',
# stream=False,
# tools=[
# PromptMessageTool(
# name='get_current_weather',
# description='Get the current weather in a given location',
# parameters={
# "type": "object",
# "properties": {
# "location": {
# "type": "string",
# "description": "The city and state e.g. San Francisco, CA"
# },
# "unit": {
# "type": "string",
# "enum": [
# "c",
# "f"
# ]
# }
# },
# "required": [
# "location"
# ]
# }
# )
# ]
# )
# assert isinstance(response, LLMResult)
# assert len(response.message.content) > 0
# assert response.usage.total_tokens > 0
# assert response.message.tool_calls[0].function.name == 'get_current_weather'
@pytest.mark.parametrize(("setup_openai_mock", "setup_xinference_mock"), [("completion", "none")], indirect=True)
def test_validate_credentials_for_generation_model(setup_openai_mock, setup_xinference_mock):
model = XinferenceAILargeLanguageModel()
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(
model="alapaca",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": "www " + os.environ.get("XINFERENCE_GENERATION_MODEL_UID"),
},
)
with pytest.raises(CredentialsValidateFailedError):
model.validate_credentials(model="alapaca", credentials={"server_url": "", "model_uid": ""})
model.validate_credentials(
model="alapaca",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_GENERATION_MODEL_UID"),
},
)
@pytest.mark.parametrize(("setup_openai_mock", "setup_xinference_mock"), [("completion", "none")], indirect=True)
def test_invoke_generation_model(setup_openai_mock, setup_xinference_mock):
model = XinferenceAILargeLanguageModel()
response = model.invoke(
model="alapaca",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_GENERATION_MODEL_UID"),
},
prompt_messages=[UserPromptMessage(content="the United States is")],
model_parameters={
"temperature": 0.7,
"top_p": 1.0,
},
stop=["you"],
user="abc-123",
stream=False,
)
assert isinstance(response, LLMResult)
assert len(response.message.content) > 0
assert response.usage.total_tokens > 0
@pytest.mark.parametrize(("setup_openai_mock", "setup_xinference_mock"), [("completion", "none")], indirect=True)
def test_invoke_stream_generation_model(setup_openai_mock, setup_xinference_mock):
model = XinferenceAILargeLanguageModel()
response = model.invoke(
model="alapaca",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_GENERATION_MODEL_UID"),
},
prompt_messages=[UserPromptMessage(content="the United States is")],
model_parameters={
"temperature": 0.7,
"top_p": 1.0,
},
stop=["you"],
stream=True,
user="abc-123",
)
assert isinstance(response, Generator)
for chunk in response:
assert isinstance(chunk, LLMResultChunk)
assert isinstance(chunk.delta, LLMResultChunkDelta)
assert isinstance(chunk.delta.message, AssistantPromptMessage)
assert len(chunk.delta.message.content) > 0 if chunk.delta.finish_reason is None else True
def test_get_num_tokens():
model = XinferenceAILargeLanguageModel()
num_tokens = model.get_num_tokens(
model="ChatGLM3",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_GENERATION_MODEL_UID"),
},
prompt_messages=[
SystemPromptMessage(
content="You are a helpful AI assistant.",
),
UserPromptMessage(content="Hello World!"),
],
tools=[
PromptMessageTool(
name="get_current_weather",
description="Get the current weather in a given location",
parameters={
"type": "object",
"properties": {
"location": {"type": "string", "description": "The city and state e.g. San Francisco, CA"},
"unit": {"type": "string", "enum": ["c", "f"]},
},
"required": ["location"],
},
)
],
)
assert isinstance(num_tokens, int)
assert num_tokens == 77
num_tokens = model.get_num_tokens(
model="ChatGLM3",
credentials={
"server_url": os.environ.get("XINFERENCE_SERVER_URL"),
"model_uid": os.environ.get("XINFERENCE_GENERATION_MODEL_UID"),
},
prompt_messages=[
SystemPromptMessage(
content="You are a helpful AI assistant.",
),
UserPromptMessage(content="Hello World!"),
],
)
assert isinstance(num_tokens, int)
assert num_tokens == 21
|