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

Merge pull request #16779 from mahenning/fix--clean-unload-embed/reranker-models

Browse Source 
Fix: Free VRAM memory when updating embedding / reranking models
This commit is contained in:
Tim Jaeryang Baek
2025-08-21 21:38:37 +04:00
committed by GitHub
parent 578f116ebb f2e78d7940
commit 5a66f69460
2 changed files with 53 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/open_webui/main.py
+13 -8
View File
@@ -924,14 +924,19 @@ try:
app.state.config.RAG_EMBEDDING_MODEL,
RAG_EMBEDDING_MODEL_AUTO_UPDATE,
)
app.state.rf = get_rf(
app.state.config.RAG_RERANKING_ENGINE,
app.state.config.RAG_RERANKING_MODEL,
app.state.config.RAG_EXTERNAL_RERANKER_URL,
app.state.config.RAG_EXTERNAL_RERANKER_API_KEY,
RAG_RERANKING_MODEL_AUTO_UPDATE,
)
if (
app.state.config.ENABLE_RAG_HYBRID_SEARCH
and not app.state.config.BYPASS_EMBEDDING_AND_RETRIEVAL
):
app.state.rf = get_rf(
app.state.config.RAG_RERANKING_ENGINE,
app.state.config.RAG_RERANKING_MODEL,
app.state.config.RAG_EXTERNAL_RERANKER_URL,
app.state.config.RAG_EXTERNAL_RERANKER_API_KEY,
RAG_RERANKING_MODEL_AUTO_UPDATE,
)
else:
app.state.rf = None
except Exception as e:
log.error(f"Error updating models: {e}")
pass