Spaces:
Build error
Build error
creating openai compatible server
Browse files- TurtleSoupBaseline/openai_api_server.py +71 -82
- requirements.txt +2 -0
TurtleSoupBaseline/openai_api_server.py
CHANGED
@@ -17,7 +17,9 @@ from transformers import AutoTokenizer, LogitsProcessor
|
|
17 |
from sse_starlette.sse import EventSourceResponse
|
18 |
|
19 |
EventSourceResponse.DEFAULT_PING_INTERVAL = 1000
|
20 |
-
MODEL_PATH =
|
|
|
|
|
21 |
MAX_MODEL_LENGTH = 8192
|
22 |
|
23 |
|
@@ -125,14 +127,16 @@ class ChatCompletionResponse(BaseModel):
|
|
125 |
model: str
|
126 |
id: str
|
127 |
object: Literal["chat.completion", "chat.completion.chunk"]
|
128 |
-
choices: List[
|
|
|
|
|
129 |
created: Optional[int] = Field(default_factory=lambda: int(time.time()))
|
130 |
usage: Optional[UsageInfo] = None
|
131 |
|
132 |
|
133 |
class InvalidScoreLogitsProcessor(LogitsProcessor):
|
134 |
def __call__(
|
135 |
-
|
136 |
) -> torch.FloatTensor:
|
137 |
if torch.isnan(scores).any() or torch.isinf(scores).any():
|
138 |
scores.zero_()
|
@@ -154,13 +158,10 @@ def process_response(output: str, use_tool: bool = False) -> Union[str, dict]:
|
|
154 |
parameters = eval(content.strip())
|
155 |
content = {
|
156 |
"name": metadata.strip(),
|
157 |
-
"arguments": json.dumps(parameters, ensure_ascii=False)
|
158 |
}
|
159 |
else:
|
160 |
-
content = {
|
161 |
-
"name": metadata.strip(),
|
162 |
-
"content": content
|
163 |
-
}
|
164 |
return content
|
165 |
|
166 |
|
@@ -174,7 +175,9 @@ async def generate_stream_glm4(params):
|
|
174 |
top_p = float(params.get("top_p", 1.0))
|
175 |
max_new_tokens = int(params.get("max_tokens", 8192))
|
176 |
messages = process_messages(messages, tools=tools, tool_choice=tool_choice)
|
177 |
-
inputs = tokenizer.apply_chat_template(
|
|
|
|
|
178 |
params_dict = {
|
179 |
"n": 1,
|
180 |
"best_of": 1,
|
@@ -195,7 +198,9 @@ async def generate_stream_glm4(params):
|
|
195 |
"skip_special_tokens": True,
|
196 |
}
|
197 |
sampling_params = SamplingParams(**params_dict)
|
198 |
-
async for output in engine.generate(
|
|
|
|
|
199 |
output_len = len(output.outputs[0].token_ids)
|
200 |
input_len = len(output.prompt_token_ids)
|
201 |
ret = {
|
@@ -203,7 +208,7 @@ async def generate_stream_glm4(params):
|
|
203 |
"usage": {
|
204 |
"prompt_tokens": input_len,
|
205 |
"completion_tokens": output_len,
|
206 |
-
"total_tokens": output_len + input_len
|
207 |
},
|
208 |
"finish_reason": output.outputs[0].finish_reason,
|
209 |
}
|
@@ -218,12 +223,13 @@ def process_messages(messages, tools=None, tool_choice="none"):
|
|
218 |
msg_has_sys = False
|
219 |
|
220 |
def filter_tools(tool_choice, tools):
|
221 |
-
function_name = tool_choice.get(
|
222 |
if not function_name:
|
223 |
return []
|
224 |
filtered_tools = [
|
225 |
-
tool
|
226 |
-
|
|
|
227 |
]
|
228 |
return filtered_tools
|
229 |
|
@@ -231,13 +237,7 @@ def process_messages(messages, tools=None, tool_choice="none"):
|
|
231 |
if isinstance(tool_choice, dict):
|
232 |
tools = filter_tools(tool_choice, tools)
|
233 |
if tools:
|
234 |
-
messages.append(
|
235 |
-
{
|
236 |
-
"role": "system",
|
237 |
-
"content": None,
|
238 |
-
"tools": tools
|
239 |
-
}
|
240 |
-
)
|
241 |
msg_has_sys = True
|
242 |
|
243 |
# add to metadata
|
@@ -246,19 +246,14 @@ def process_messages(messages, tools=None, tool_choice="none"):
|
|
246 |
{
|
247 |
"role": "assistant",
|
248 |
"metadata": tool_choice["function"]["name"],
|
249 |
-
"content": ""
|
250 |
}
|
251 |
)
|
252 |
|
253 |
for m in _messages:
|
254 |
role, content, func_call = m.role, m.content, m.function_call
|
255 |
if role == "function":
|
256 |
-
messages.append(
|
257 |
-
{
|
258 |
-
"role": "observation",
|
259 |
-
"content": content
|
260 |
-
}
|
261 |
-
)
|
262 |
elif role == "assistant" and func_call is not None:
|
263 |
for response in content.split("<|assistant|>"):
|
264 |
if "\n" in response:
|
@@ -266,11 +261,7 @@ def process_messages(messages, tools=None, tool_choice="none"):
|
|
266 |
else:
|
267 |
metadata, sub_content = "", response
|
268 |
messages.append(
|
269 |
-
{
|
270 |
-
"role": role,
|
271 |
-
"metadata": metadata,
|
272 |
-
"content": sub_content.strip()
|
273 |
-
}
|
274 |
)
|
275 |
else:
|
276 |
if role == "system" and msg_has_sys:
|
@@ -315,7 +306,9 @@ async def create_chat_completion(request: ChatCompletionRequest):
|
|
315 |
predict_stream_generator = predict_stream(request.model, gen_params)
|
316 |
output = await anext(predict_stream_generator)
|
317 |
if output:
|
318 |
-
return EventSourceResponse(
|
|
|
|
|
319 |
logger.debug(f"First result output:\n{output}")
|
320 |
|
321 |
function_call = None
|
@@ -332,7 +325,9 @@ async def create_chat_completion(request: ChatCompletionRequest):
|
|
332 |
if not gen_params.get("messages"):
|
333 |
gen_params["messages"] = []
|
334 |
gen_params["messages"].append(ChatMessage(role="assistant", content=output))
|
335 |
-
gen_params["messages"].append(
|
|
|
|
|
336 |
generate = predict(request.model, gen_params)
|
337 |
return EventSourceResponse(generate, media_type="text/event-stream")
|
338 |
else:
|
@@ -354,7 +349,8 @@ async def create_chat_completion(request: ChatCompletionRequest):
|
|
354 |
function_call = process_response(response["text"], use_tool=True)
|
355 |
except:
|
356 |
logger.warning(
|
357 |
-
"Failed to parse tool call, maybe the response is not a function call(such as cogview drawing) or have been answered."
|
|
|
358 |
|
359 |
if isinstance(function_call, dict):
|
360 |
finish_reason = "function_call"
|
@@ -363,7 +359,9 @@ async def create_chat_completion(request: ChatCompletionRequest):
|
|
363 |
message = ChatMessage(
|
364 |
role="assistant",
|
365 |
content=response["text"],
|
366 |
-
function_call=
|
|
|
|
|
367 |
)
|
368 |
|
369 |
logger.debug(f"==== message ====\n{message}")
|
@@ -382,23 +380,23 @@ async def create_chat_completion(request: ChatCompletionRequest):
|
|
382 |
id="", # for open_source model, id is empty
|
383 |
choices=[choice_data],
|
384 |
object="chat.completion",
|
385 |
-
usage=usage
|
386 |
)
|
387 |
|
388 |
|
389 |
async def predict(model_id: str, params: dict):
|
390 |
choice_data = ChatCompletionResponseStreamChoice(
|
391 |
-
index=0,
|
392 |
-
|
393 |
-
|
|
|
394 |
)
|
395 |
-
chunk = ChatCompletionResponse(model=model_id, id="", choices=[choice_data], object="chat.completion.chunk")
|
396 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
397 |
|
398 |
previous_text = ""
|
399 |
async for new_response in generate_stream_glm4(params):
|
400 |
decoded_unicode = new_response["text"]
|
401 |
-
delta_text = decoded_unicode[len(previous_text):]
|
402 |
previous_text = decoded_unicode
|
403 |
|
404 |
finish_reason = new_response["finish_reason"]
|
@@ -411,7 +409,8 @@ async def predict(model_id: str, params: dict):
|
|
411 |
function_call = process_response(decoded_unicode, use_tool=True)
|
412 |
except:
|
413 |
logger.warning(
|
414 |
-
"Failed to parse tool call, maybe the response is not a tool call or have been answered."
|
|
|
415 |
|
416 |
if isinstance(function_call, dict):
|
417 |
function_call = FunctionCallResponse(**function_call)
|
@@ -419,48 +418,42 @@ async def predict(model_id: str, params: dict):
|
|
419 |
delta = DeltaMessage(
|
420 |
content=delta_text,
|
421 |
role="assistant",
|
422 |
-
function_call=
|
|
|
|
|
|
|
|
|
423 |
)
|
424 |
|
425 |
choice_data = ChatCompletionResponseStreamChoice(
|
426 |
-
index=0,
|
427 |
-
delta=delta,
|
428 |
-
finish_reason=finish_reason
|
429 |
)
|
430 |
chunk = ChatCompletionResponse(
|
431 |
-
model=model_id,
|
432 |
-
id="",
|
433 |
-
choices=[choice_data],
|
434 |
-
object="chat.completion.chunk"
|
435 |
)
|
436 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
437 |
|
438 |
choice_data = ChatCompletionResponseStreamChoice(
|
439 |
-
index=0,
|
440 |
-
delta=DeltaMessage(),
|
441 |
-
finish_reason="stop"
|
442 |
)
|
443 |
chunk = ChatCompletionResponse(
|
444 |
-
model=model_id,
|
445 |
-
id="",
|
446 |
-
choices=[choice_data],
|
447 |
-
object="chat.completion.chunk"
|
448 |
)
|
449 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
450 |
-
yield
|
451 |
|
452 |
|
453 |
async def predict_stream(model_id, gen_params):
|
454 |
output = ""
|
455 |
is_function_call = False
|
456 |
has_send_first_chunk = False
|
457 |
-
async
|
458 |
decoded_unicode = new_response["text"]
|
459 |
-
delta_text = decoded_unicode[len(output):]
|
460 |
output = decoded_unicode
|
461 |
|
462 |
if not is_function_call and len(output) > 7:
|
463 |
-
is_function_call = output and
|
464 |
if is_function_call:
|
465 |
continue
|
466 |
|
@@ -472,16 +465,14 @@ async def predict_stream(model_id, gen_params):
|
|
472 |
function_call=None,
|
473 |
)
|
474 |
choice_data = ChatCompletionResponseStreamChoice(
|
475 |
-
index=0,
|
476 |
-
delta=message,
|
477 |
-
finish_reason=finish_reason
|
478 |
)
|
479 |
chunk = ChatCompletionResponse(
|
480 |
model=model_id,
|
481 |
id="",
|
482 |
choices=[choice_data],
|
483 |
created=int(time.time()),
|
484 |
-
object="chat.completion.chunk"
|
485 |
)
|
486 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
487 |
|
@@ -493,41 +484,39 @@ async def predict_stream(model_id, gen_params):
|
|
493 |
function_call=None,
|
494 |
)
|
495 |
choice_data = ChatCompletionResponseStreamChoice(
|
496 |
-
index=0,
|
497 |
-
delta=message,
|
498 |
-
finish_reason=finish_reason
|
499 |
)
|
500 |
chunk = ChatCompletionResponse(
|
501 |
model=model_id,
|
502 |
id="",
|
503 |
choices=[choice_data],
|
504 |
created=int(time.time()),
|
505 |
-
object="chat.completion.chunk"
|
506 |
)
|
507 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
508 |
|
509 |
if is_function_call:
|
510 |
yield output
|
511 |
else:
|
512 |
-
yield
|
513 |
|
514 |
|
515 |
async def parse_output_text(model_id: str, value: str):
|
516 |
choice_data = ChatCompletionResponseStreamChoice(
|
517 |
-
index=0,
|
518 |
-
|
519 |
-
|
|
|
520 |
)
|
521 |
-
chunk = ChatCompletionResponse(model=model_id, id="", choices=[choice_data], object="chat.completion.chunk")
|
522 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
523 |
choice_data = ChatCompletionResponseStreamChoice(
|
524 |
-
index=0,
|
525 |
-
|
526 |
-
|
|
|
527 |
)
|
528 |
-
chunk = ChatCompletionResponse(model=model_id, id="", choices=[choice_data], object="chat.completion.chunk")
|
529 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
530 |
-
yield
|
531 |
|
532 |
|
533 |
if __name__ == "__main__":
|
@@ -546,4 +535,4 @@ if __name__ == "__main__":
|
|
546 |
max_model_len=MAX_MODEL_LENGTH,
|
547 |
)
|
548 |
engine = AsyncLLMEngine.from_engine_args(engine_args)
|
549 |
-
uvicorn.run(app, host=
|
|
|
17 |
from sse_starlette.sse import EventSourceResponse
|
18 |
|
19 |
EventSourceResponse.DEFAULT_PING_INTERVAL = 1000
|
20 |
+
MODEL_PATH = (
|
21 |
+
"../llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528"
|
22 |
+
)
|
23 |
MAX_MODEL_LENGTH = 8192
|
24 |
|
25 |
|
|
|
127 |
model: str
|
128 |
id: str
|
129 |
object: Literal["chat.completion", "chat.completion.chunk"]
|
130 |
+
choices: List[
|
131 |
+
Union[ChatCompletionResponseChoice, ChatCompletionResponseStreamChoice]
|
132 |
+
]
|
133 |
created: Optional[int] = Field(default_factory=lambda: int(time.time()))
|
134 |
usage: Optional[UsageInfo] = None
|
135 |
|
136 |
|
137 |
class InvalidScoreLogitsProcessor(LogitsProcessor):
|
138 |
def __call__(
|
139 |
+
self, input_ids: torch.LongTensor, scores: torch.FloatTensor
|
140 |
) -> torch.FloatTensor:
|
141 |
if torch.isnan(scores).any() or torch.isinf(scores).any():
|
142 |
scores.zero_()
|
|
|
158 |
parameters = eval(content.strip())
|
159 |
content = {
|
160 |
"name": metadata.strip(),
|
161 |
+
"arguments": json.dumps(parameters, ensure_ascii=False),
|
162 |
}
|
163 |
else:
|
164 |
+
content = {"name": metadata.strip(), "content": content}
|
|
|
|
|
|
|
165 |
return content
|
166 |
|
167 |
|
|
|
175 |
top_p = float(params.get("top_p", 1.0))
|
176 |
max_new_tokens = int(params.get("max_tokens", 8192))
|
177 |
messages = process_messages(messages, tools=tools, tool_choice=tool_choice)
|
178 |
+
inputs = tokenizer.apply_chat_template(
|
179 |
+
messages, add_generation_prompt=True, tokenize=False
|
180 |
+
)
|
181 |
params_dict = {
|
182 |
"n": 1,
|
183 |
"best_of": 1,
|
|
|
198 |
"skip_special_tokens": True,
|
199 |
}
|
200 |
sampling_params = SamplingParams(**params_dict)
|
201 |
+
async for output in engine.generate(
|
202 |
+
inputs=inputs, sampling_params=sampling_params, request_id="glm-4-9b"
|
203 |
+
):
|
204 |
output_len = len(output.outputs[0].token_ids)
|
205 |
input_len = len(output.prompt_token_ids)
|
206 |
ret = {
|
|
|
208 |
"usage": {
|
209 |
"prompt_tokens": input_len,
|
210 |
"completion_tokens": output_len,
|
211 |
+
"total_tokens": output_len + input_len,
|
212 |
},
|
213 |
"finish_reason": output.outputs[0].finish_reason,
|
214 |
}
|
|
|
223 |
msg_has_sys = False
|
224 |
|
225 |
def filter_tools(tool_choice, tools):
|
226 |
+
function_name = tool_choice.get("function", {}).get("name", None)
|
227 |
if not function_name:
|
228 |
return []
|
229 |
filtered_tools = [
|
230 |
+
tool
|
231 |
+
for tool in tools
|
232 |
+
if tool.get("function", {}).get("name") == function_name
|
233 |
]
|
234 |
return filtered_tools
|
235 |
|
|
|
237 |
if isinstance(tool_choice, dict):
|
238 |
tools = filter_tools(tool_choice, tools)
|
239 |
if tools:
|
240 |
+
messages.append({"role": "system", "content": None, "tools": tools})
|
|
|
|
|
|
|
|
|
|
|
|
|
241 |
msg_has_sys = True
|
242 |
|
243 |
# add to metadata
|
|
|
246 |
{
|
247 |
"role": "assistant",
|
248 |
"metadata": tool_choice["function"]["name"],
|
249 |
+
"content": "",
|
250 |
}
|
251 |
)
|
252 |
|
253 |
for m in _messages:
|
254 |
role, content, func_call = m.role, m.content, m.function_call
|
255 |
if role == "function":
|
256 |
+
messages.append({"role": "observation", "content": content})
|
|
|
|
|
|
|
|
|
|
|
257 |
elif role == "assistant" and func_call is not None:
|
258 |
for response in content.split("<|assistant|>"):
|
259 |
if "\n" in response:
|
|
|
261 |
else:
|
262 |
metadata, sub_content = "", response
|
263 |
messages.append(
|
264 |
+
{"role": role, "metadata": metadata, "content": sub_content.strip()}
|
|
|
|
|
|
|
|
|
265 |
)
|
266 |
else:
|
267 |
if role == "system" and msg_has_sys:
|
|
|
306 |
predict_stream_generator = predict_stream(request.model, gen_params)
|
307 |
output = await anext(predict_stream_generator)
|
308 |
if output:
|
309 |
+
return EventSourceResponse(
|
310 |
+
predict_stream_generator, media_type="text/event-stream"
|
311 |
+
)
|
312 |
logger.debug(f"First result output:\n{output}")
|
313 |
|
314 |
function_call = None
|
|
|
325 |
if not gen_params.get("messages"):
|
326 |
gen_params["messages"] = []
|
327 |
gen_params["messages"].append(ChatMessage(role="assistant", content=output))
|
328 |
+
gen_params["messages"].append(
|
329 |
+
ChatMessage(role="tool", name=function_call.name, content=tool_response)
|
330 |
+
)
|
331 |
generate = predict(request.model, gen_params)
|
332 |
return EventSourceResponse(generate, media_type="text/event-stream")
|
333 |
else:
|
|
|
349 |
function_call = process_response(response["text"], use_tool=True)
|
350 |
except:
|
351 |
logger.warning(
|
352 |
+
"Failed to parse tool call, maybe the response is not a function call(such as cogview drawing) or have been answered."
|
353 |
+
)
|
354 |
|
355 |
if isinstance(function_call, dict):
|
356 |
finish_reason = "function_call"
|
|
|
359 |
message = ChatMessage(
|
360 |
role="assistant",
|
361 |
content=response["text"],
|
362 |
+
function_call=(
|
363 |
+
function_call if isinstance(function_call, FunctionCallResponse) else None
|
364 |
+
),
|
365 |
)
|
366 |
|
367 |
logger.debug(f"==== message ====\n{message}")
|
|
|
380 |
id="", # for open_source model, id is empty
|
381 |
choices=[choice_data],
|
382 |
object="chat.completion",
|
383 |
+
usage=usage,
|
384 |
)
|
385 |
|
386 |
|
387 |
async def predict(model_id: str, params: dict):
|
388 |
choice_data = ChatCompletionResponseStreamChoice(
|
389 |
+
index=0, delta=DeltaMessage(role="assistant"), finish_reason=None
|
390 |
+
)
|
391 |
+
chunk = ChatCompletionResponse(
|
392 |
+
model=model_id, id="", choices=[choice_data], object="chat.completion.chunk"
|
393 |
)
|
|
|
394 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
395 |
|
396 |
previous_text = ""
|
397 |
async for new_response in generate_stream_glm4(params):
|
398 |
decoded_unicode = new_response["text"]
|
399 |
+
delta_text = decoded_unicode[len(previous_text) :]
|
400 |
previous_text = decoded_unicode
|
401 |
|
402 |
finish_reason = new_response["finish_reason"]
|
|
|
409 |
function_call = process_response(decoded_unicode, use_tool=True)
|
410 |
except:
|
411 |
logger.warning(
|
412 |
+
"Failed to parse tool call, maybe the response is not a tool call or have been answered."
|
413 |
+
)
|
414 |
|
415 |
if isinstance(function_call, dict):
|
416 |
function_call = FunctionCallResponse(**function_call)
|
|
|
418 |
delta = DeltaMessage(
|
419 |
content=delta_text,
|
420 |
role="assistant",
|
421 |
+
function_call=(
|
422 |
+
function_call
|
423 |
+
if isinstance(function_call, FunctionCallResponse)
|
424 |
+
else None
|
425 |
+
),
|
426 |
)
|
427 |
|
428 |
choice_data = ChatCompletionResponseStreamChoice(
|
429 |
+
index=0, delta=delta, finish_reason=finish_reason
|
|
|
|
|
430 |
)
|
431 |
chunk = ChatCompletionResponse(
|
432 |
+
model=model_id, id="", choices=[choice_data], object="chat.completion.chunk"
|
|
|
|
|
|
|
433 |
)
|
434 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
435 |
|
436 |
choice_data = ChatCompletionResponseStreamChoice(
|
437 |
+
index=0, delta=DeltaMessage(), finish_reason="stop"
|
|
|
|
|
438 |
)
|
439 |
chunk = ChatCompletionResponse(
|
440 |
+
model=model_id, id="", choices=[choice_data], object="chat.completion.chunk"
|
|
|
|
|
|
|
441 |
)
|
442 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
443 |
+
yield "[DONE]"
|
444 |
|
445 |
|
446 |
async def predict_stream(model_id, gen_params):
|
447 |
output = ""
|
448 |
is_function_call = False
|
449 |
has_send_first_chunk = False
|
450 |
+
async for new_response in generate_stream_glm4(gen_params):
|
451 |
decoded_unicode = new_response["text"]
|
452 |
+
delta_text = decoded_unicode[len(output) :]
|
453 |
output = decoded_unicode
|
454 |
|
455 |
if not is_function_call and len(output) > 7:
|
456 |
+
is_function_call = output and "get_" in output
|
457 |
if is_function_call:
|
458 |
continue
|
459 |
|
|
|
465 |
function_call=None,
|
466 |
)
|
467 |
choice_data = ChatCompletionResponseStreamChoice(
|
468 |
+
index=0, delta=message, finish_reason=finish_reason
|
|
|
|
|
469 |
)
|
470 |
chunk = ChatCompletionResponse(
|
471 |
model=model_id,
|
472 |
id="",
|
473 |
choices=[choice_data],
|
474 |
created=int(time.time()),
|
475 |
+
object="chat.completion.chunk",
|
476 |
)
|
477 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
478 |
|
|
|
484 |
function_call=None,
|
485 |
)
|
486 |
choice_data = ChatCompletionResponseStreamChoice(
|
487 |
+
index=0, delta=message, finish_reason=finish_reason
|
|
|
|
|
488 |
)
|
489 |
chunk = ChatCompletionResponse(
|
490 |
model=model_id,
|
491 |
id="",
|
492 |
choices=[choice_data],
|
493 |
created=int(time.time()),
|
494 |
+
object="chat.completion.chunk",
|
495 |
)
|
496 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
497 |
|
498 |
if is_function_call:
|
499 |
yield output
|
500 |
else:
|
501 |
+
yield "[DONE]"
|
502 |
|
503 |
|
504 |
async def parse_output_text(model_id: str, value: str):
|
505 |
choice_data = ChatCompletionResponseStreamChoice(
|
506 |
+
index=0, delta=DeltaMessage(role="assistant", content=value), finish_reason=None
|
507 |
+
)
|
508 |
+
chunk = ChatCompletionResponse(
|
509 |
+
model=model_id, id="", choices=[choice_data], object="chat.completion.chunk"
|
510 |
)
|
|
|
511 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
512 |
choice_data = ChatCompletionResponseStreamChoice(
|
513 |
+
index=0, delta=DeltaMessage(), finish_reason="stop"
|
514 |
+
)
|
515 |
+
chunk = ChatCompletionResponse(
|
516 |
+
model=model_id, id="", choices=[choice_data], object="chat.completion.chunk"
|
517 |
)
|
|
|
518 |
yield "{}".format(chunk.model_dump_json(exclude_unset=True))
|
519 |
+
yield "[DONE]"
|
520 |
|
521 |
|
522 |
if __name__ == "__main__":
|
|
|
535 |
max_model_len=MAX_MODEL_LENGTH,
|
536 |
)
|
537 |
engine = AsyncLLMEngine.from_engine_args(engine_args)
|
538 |
+
uvicorn.run(app, host="0.0.0.0", port=8000, workers=1)
|
requirements.txt
CHANGED
@@ -14,3 +14,5 @@ langchain_openai==0.1.13
|
|
14 |
wandb==0.17.4
|
15 |
# triton
|
16 |
# xformers
|
|
|
|
|
|
14 |
wandb==0.17.4
|
15 |
# triton
|
16 |
# xformers
|
17 |
+
uvicorn
|
18 |
+
vllm
|