gardel/lightrag
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Use the context manager for the openai client

Browse Source 
This avoids issues of resource cleanup (too many open files) when dealing with massively parallel calls to the openai API since RAII in python is highly unreliable in such contexts.
This commit is contained in:
Arjun Rao
2025-05-08 11:42:53 +10:00
parent f2c522ce7a
commit b7eae4d7c0
1 changed files with 14 additions and 12 deletions
Download Patch File Download Diff File Expand all files Collapse all files

26
lightrag/llm/openai.py
View File

@@ -177,14 +177,15 @@ async def openai_complete_if_cache(
logger.debug("===== Sending Query to LLM =====")
try:
if "response_format" in kwargs:
response = await openai_async_client.beta.chat.completions.parse(
model=model, messages=messages, **kwargs
)
else:
response = await openai_async_client.chat.completions.create(
model=model, messages=messages, **kwargs
)
async with openai_async_client:
if "response_format" in kwargs:
response = await openai_async_client.beta.chat.completions.parse(
model=model, messages=messages, **kwargs
)
else:
response = await openai_async_client.chat.completions.create(
model=model, messages=messages, **kwargs
)
except APIConnectionError as e:
logger.error(f"OpenAI API Connection Error: {e}")
raise
@@ -421,7 +422,8 @@ async def openai_embed(
api_key=api_key, base_url=base_url, client_configs=client_configs
)
response = await openai_async_client.embeddings.create(
model=model, input=texts, encoding_format="float"
)
return np.array([dp.embedding for dp in response.data])
async with openai_async_client:
response = await openai_async_client.embeddings.create(
model=model, input=texts, encoding_format="float"
)
return np.array([dp.embedding for dp in response.data])