Spaces:
Sleeping
Sleeping
Commit History
-
fix: format history as string to prevent Llama 3 hallucinating assistant role during contextualization
Dus Trance8ae9c -
fix: prevent LLM from hallucinating queries during contextualization
Dus Tran428313f -
fix: preload models on startup and increase LLM timeout
Dus Tran9a01e3a -
fix: soften strict prompt for keyword queries
Dus Tran3ff2322 -
fix: handle 2D numpy array from multilingual reranker
Dus Tranbd6c976 -
fix: switch to multilingual reranker
Dus Trand6fdc60 -
fix: dockerignore and memory init
Dus Tranc8a642a -
chore: remove IS_RENDER and gemini imports
Dus Tranfbac3e6
-
feat(rag): enable rerank model on render
Dus Tran463f005 -
feat(rag): use bkai huggingface model universally
Dus Tran0bbf4ee -
fix(prompt): allow LLM to respond to chit-chat naturally
Dus Tran3f20a80 -
fix(rag): update prompt and reduce temperature to prevent hallucinations
Dus Tran3380923 -
fix: use models/gemini-embedding-2 with 768 dimensions for compatibility with local database
Dus Trandc6ff05 -
fix: correct GoogleGenerativeAIEmbeddings class name
Dus Tranf475e72 -
fix: make heavy ML libraries import conditionally on Render to prevent OOM/timeouts
Dus Tranaf847aa -
fix: strip carriage returns from database url
Dus Tran525c11d -
feat: tích hợp supabase database và cấu hình github actions ci
Dus Tranea19adc
-
“…I”
Dus Tranfeb5410
-
“…I”
Dus Trancf573aa -
hoàn thiện Multimodal RAG chatbot với FastAPI Backend & Web UI
Dus Tran44ed2c7