vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2025-09-10 15:29:39 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

add check-para

Browse source
This commit is contained in:
Alisehen 2025-04-22 09:30:08 +00:00
parent 485588017b
commit c995bdbbfa
2 changed files with 114 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

54
ktransformers/server/api/openai/legacy/completions.py
View file

@ -7,13 +7,63 @@ from ktransformers.server.utils.create_interface import get_interface
from ktransformers.server.schemas.assistants.streaming import stream_response
from ktransformers.server.schemas.legacy.completions import CompletionCreate,CompletionObject
from ktransformers.server.schemas.endpoints.chat import RawUsage
from fastapi.responses import JSONResponse
from ktransformers.server.config.config import Config
router = APIRouter()
@router.post("/completions",tags=['openai'])
async def create_completion(request:Request, create:CompletionCreate):
id = str(uuid4())
if create.model != Config().model_name:
return JSONResponse(
status_code=400,
content={
"error": {
"message": "Model not found",
"code": 404,
"type": "NotFound"
}
})
if create.max_tokens<0:
return JSONResponse(
status_code=400,
content={
"object": "error",
"message": f"max_new_tokens must be at least 0, got {create.max_tokens}.",
"type": "BadRequestError",
"param": None,
"code": 400
})
if create.max_completion_tokens<0:
return JSONResponse(
status_code=400,
content={
"object": "error",
"message": f"max_new_tokens must be at least 0, got {create.max_completion_tokens}.",
"type": "BadRequestError",
"param": None,
"code": 400
})
if create.temperature<0 or create.temperature>2:
return JSONResponse(
status_code=400,
content={
"object": "error",
"message": f"temperature must be in [0, 2], got {create.temperature}.",
"type": "BadRequestError",
"param": None,
"code": 400
})
if create.top_p<=0 or create.top_p>1:
return JSONResponse(
status_code=400,
content={
"object": "error",
"message": f"top_p must be in (0, 1], got {create.top_p}.",
"type": "BadRequestError",
"param": None,
"code": 400
})
interface = get_interface()
print(f'COMPLETION INPUT:----\n{create.prompt}\n----')