baalajimaestro/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity

fix(server): minor type fixes

Browse source
This commit is contained in:
Andrei Betlen 2024-03-23 17:14:15 -04:00
parent c1325dcdfb
commit d11ccc3036
2 changed files with 4 additions and 4 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
llama_cpp/server/app.py
View file

@ -493,7 +493,7 @@ async def tokenize(
) -> TokenizeInputResponse:
tokens = llama_proxy(body.model).tokenize(body.input.encode("utf-8"), special=True)
return {"tokens": tokens}
return TokenizeInputResponse(tokens=tokens)
@router.post(
@ -508,7 +508,7 @@ async def count_query_tokens(
) -> TokenizeInputCountResponse:
tokens = llama_proxy(body.model).tokenize(body.input.encode("utf-8"), special=True)
return {"count": len(tokens)}
return TokenizeInputCountResponse(count=len(tokens))
@router.post(
@ -523,4 +523,4 @@ async def detokenize(
) -> DetokenizeInputResponse:
text = llama_proxy(body.model).detokenize(body.tokens).decode("utf-8")
return {"text": text}
return DetokenizeInputResponse(text=text)

2
llama_cpp/server/types.py
View file

@ -268,7 +268,7 @@ class ModelList(TypedDict):
class TokenizeInputRequest(BaseModel):
model: Optional[str] = model_field
input: Optional[str] = Field(description="The input to tokenize.")
input: str = Field(description="The input to tokenize.")
model_config = {
"json_schema_extra": {"examples": [{"input": "How many tokens in this query?"}]}