github/gpt4all
1
0
Fork 0
mirror of https://github.com/nomic-ai/gpt4all.git synced 2025-08-06 10:33:38 +00:00
Code Issues Packages Projects Releases Wiki Activity

Moved model check into cpu only paths

Browse Source
This commit is contained in:
Andriy Mulyar 2023-07-24 11:34:50 -04:00
parent 8aba2c9009
commit 3d10110314
1 changed files with 4 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
gpt4all-api/gpt4all_api/app/api_v1/routes/completions.py
View File

@ -111,10 +111,6 @@ async def completions(request: CompletionRequest):
''' '''
Completes a GPT4All model response. Completes a GPT4All model response.
''' '''
if request.model != settings.model:
raise HTTPException(status_code=400, detail=f"The GPT4All inference server is booted to only infer: `{settings.model}`")
if settings.inference_mode == "gpu": if settings.inference_mode == "gpu":
params = request.dict(exclude={'model', 'prompt', 'max_tokens', 'n'}) params = request.dict(exclude={'model', 'prompt', 'max_tokens', 'n'})
params["max_new_tokens"] = request.max_tokens params["max_new_tokens"] = request.max_tokens
@ -170,6 +166,10 @@ async def completions(request: CompletionRequest):
else: else:
if request.model != settings.model:
raise HTTPException(status_code=400,
detail=f"The GPT4All inference server is booted to only infer: `{settings.model}`")
if isinstance(request.prompt, list): if isinstance(request.prompt, list):
if len(request.prompt) > 1: if len(request.prompt) > 1:
raise HTTPException(status_code=400, detail="Can only infer one inference per request in CPU mode.") raise HTTPException(status_code=400, detail="Can only infer one inference per request in CPU mode.")