Emil_Shanaty/open-webui-ai4me
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

enh: RAG_EMBEDDING_CONCURRENT_REQUESTS

Browse Source
This commit is contained in:
Timothy Jaeryang Baek
2026-02-21 14:33:48 -06:00
parent 5522b91c32
commit 5d4547f934
5 changed files with 64 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/open_webui/routers/retrieval.py
+11 -2
View File
@@ -270,6 +270,7 @@ async def get_status(request: Request):
"RAG_RERANKING_MODEL": request.app.state.config.RAG_RERANKING_MODEL,
"RAG_EMBEDDING_BATCH_SIZE": request.app.state.config.RAG_EMBEDDING_BATCH_SIZE,
"ENABLE_ASYNC_EMBEDDING": request.app.state.config.ENABLE_ASYNC_EMBEDDING,
"RAG_EMBEDDING_CONCURRENT_REQUESTS": request.app.state.config.RAG_EMBEDDING_CONCURRENT_REQUESTS,
}
@@ -281,6 +282,7 @@ async def get_embedding_config(request: Request, user=Depends(get_admin_user)):
"RAG_EMBEDDING_MODEL": request.app.state.config.RAG_EMBEDDING_MODEL,
"RAG_EMBEDDING_BATCH_SIZE": request.app.state.config.RAG_EMBEDDING_BATCH_SIZE,
"ENABLE_ASYNC_EMBEDDING": request.app.state.config.ENABLE_ASYNC_EMBEDDING,
"RAG_EMBEDDING_CONCURRENT_REQUESTS": request.app.state.config.RAG_EMBEDDING_CONCURRENT_REQUESTS,
"openai_config": {
"url": request.app.state.config.RAG_OPENAI_API_BASE_URL,
"key": request.app.state.config.RAG_OPENAI_API_KEY,
@@ -321,6 +323,7 @@ class EmbeddingModelUpdateForm(BaseModel):
RAG_EMBEDDING_MODEL: str
RAG_EMBEDDING_BATCH_SIZE: Optional[int] = 1
ENABLE_ASYNC_EMBEDDING: Optional[bool] = True
RAG_EMBEDDING_CONCURRENT_REQUESTS: Optional[int] = 0
def unload_embedding_model(request: Request):
@@ -355,6 +358,9 @@ async def update_embedding_config(
request.app.state.config.ENABLE_ASYNC_EMBEDDING = (
form_data.ENABLE_ASYNC_EMBEDDING
)
request.app.state.config.RAG_EMBEDDING_CONCURRENT_REQUESTS = (
form_data.RAG_EMBEDDING_CONCURRENT_REQUESTS
)
if request.app.state.config.RAG_EMBEDDING_ENGINE in [
"ollama",
@@ -422,6 +428,7 @@ async def update_embedding_config(
else None
),
enable_async=request.app.state.config.ENABLE_ASYNC_EMBEDDING,
concurrent_requests=request.app.state.config.RAG_EMBEDDING_CONCURRENT_REQUESTS,
)
return {
@@ -430,6 +437,7 @@ async def update_embedding_config(
"RAG_EMBEDDING_MODEL": request.app.state.config.RAG_EMBEDDING_MODEL,
"RAG_EMBEDDING_BATCH_SIZE": request.app.state.config.RAG_EMBEDDING_BATCH_SIZE,
"ENABLE_ASYNC_EMBEDDING": request.app.state.config.ENABLE_ASYNC_EMBEDDING,
"RAG_EMBEDDING_CONCURRENT_REQUESTS": request.app.state.config.RAG_EMBEDDING_CONCURRENT_REQUESTS,
"openai_config": {
"url": request.app.state.config.RAG_OPENAI_API_BASE_URL,
"key": request.app.state.config.RAG_OPENAI_API_KEY,
@@ -1589,6 +1597,7 @@ def save_docs_to_vector_db(
else None
),
enable_async=request.app.state.config.ENABLE_ASYNC_EMBEDDING,
concurrent_requests=request.app.state.config.RAG_EMBEDDING_CONCURRENT_REQUESTS,
)
# Run async embedding in sync context using the main event loop
@@ -2329,7 +2338,7 @@ async def process_web_search(
# Limited concurrency with semaphore
semaphore = asyncio.Semaphore(concurrent_limit)
async def search_with_limit(query):
async def search_query_with_semaphore(query):
async with semaphore:
return await run_in_threadpool(
search_web,
@@ -2339,7 +2348,7 @@ async def process_web_search(
user,
)
search_tasks = [search_with_limit(query) for query in form_data.queries]
search_tasks = [search_query_with_semaphore(query) for query in form_data.queries]
else:
# Unlimited parallel execution (previous behavior)
search_tasks = [