ti.sil/open-webui
1
0
Fork 0
mirror of https://github.com/open-webui/open-webui.git synced 2025-12-15 05:45:19 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

Formatting

Browse source
This commit is contained in:
Marko Henning 2025-08-21 13:40:56 +02:00
parent b3de3295d6
commit c821c3ecb0
2 changed files with 16 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

5
backend/open_webui/main.py
View file

@ -924,7 +924,10 @@ try:
app.state.config.RAG_EMBEDDING_MODEL, app.state.config.RAG_EMBEDDING_MODEL,
RAG_EMBEDDING_MODEL_AUTO_UPDATE, RAG_EMBEDDING_MODEL_AUTO_UPDATE,
) )
if app.state.config.ENABLE_RAG_HYBRID_SEARCH and not app.state.config.BYPASS_EMBEDDING_AND_RETRIEVAL: if (
app.state.config.ENABLE_RAG_HYBRID_SEARCH
and not app.state.config.BYPASS_EMBEDDING_AND_RETRIEVAL
):
app.state.rf = get_rf( app.state.rf = get_rf(
app.state.config.RAG_RERANKING_ENGINE, app.state.config.RAG_RERANKING_ENGINE,
app.state.config.RAG_RERANKING_MODEL, app.state.config.RAG_RERANKING_MODEL,

17
backend/open_webui/routers/retrieval.py
View file

@ -280,14 +280,16 @@ async def update_embedding_config(
log.info( log.info(
f"Updating embedding model: {request.app.state.config.RAG_EMBEDDING_MODEL} to {form_data.embedding_model}" f"Updating embedding model: {request.app.state.config.RAG_EMBEDDING_MODEL} to {form_data.embedding_model}"
) )
if request.app.state.config.RAG_EMBEDDING_ENGINE == '': if request.app.state.config.RAG_EMBEDDING_ENGINE == "":
# unloads current internal embedding model and clears VRAM cache # unloads current internal embedding model and clears VRAM cache
request.app.state.ef = None request.app.state.ef = None
request.app.state.EMBEDDING_FUNCTION = None request.app.state.EMBEDDING_FUNCTION = None
import gc import gc
gc.collect() gc.collect()
if DEVICE_TYPE == 'cuda': if DEVICE_TYPE == "cuda":
import torch import torch
if torch.cuda.is_available(): if torch.cuda.is_available():
torch.cuda.empty_cache() torch.cuda.empty_cache()
try: try:
@ -815,14 +817,16 @@ async def update_rag_config(
) )
# Reranking settings # Reranking settings
if request.app.state.config.RAG_RERANKING_ENGINE == '': if request.app.state.config.RAG_RERANKING_ENGINE == "":
# Unloading the internal reranker and clear VRAM memory # Unloading the internal reranker and clear VRAM memory
request.app.state.rf = None request.app.state.rf = None
request.app.state.RERANKING_FUNCTION = None request.app.state.RERANKING_FUNCTION = None
import gc import gc
gc.collect() gc.collect()
if DEVICE_TYPE == 'cuda': if DEVICE_TYPE == "cuda":
import torch import torch
if torch.cuda.is_available(): if torch.cuda.is_available():
torch.cuda.empty_cache() torch.cuda.empty_cache()
request.app.state.config.RAG_RERANKING_ENGINE = ( request.app.state.config.RAG_RERANKING_ENGINE = (
@ -854,7 +858,10 @@ async def update_rag_config(
) )
try: try:
if request.app.state.config.ENABLE_RAG_HYBRID_SEARCH and not request.app.state.config.BYPASS_EMBEDDING_AND_RETRIEVAL: if (
request.app.state.config.ENABLE_RAG_HYBRID_SEARCH
and not request.app.state.config.BYPASS_EMBEDDING_AND_RETRIEVAL
):
request.app.state.rf = get_rf( request.app.state.rf = get_rf(
request.app.state.config.RAG_RERANKING_ENGINE, request.app.state.config.RAG_RERANKING_ENGINE,
request.app.state.config.RAG_RERANKING_MODEL, request.app.state.config.RAG_RERANKING_MODEL,