docker run --detach --publish 8501:8501 --env OLLAMA_BASE_URL=http://<ollama host>:11434 startnow65/ragbase
| Name | Description | Default value |
|---|---|---|
STREAMLIT_SERVER_MAX_UPLOAD_SIZE | Maximum size of each file to upload for analysis in MB | "10" |
APP_HOME | Used as base for app related files | "/app" |
LLM_EMBEDDINGS | Model to use for Embeddings | "BAAI/bge-base-en-v1.5" |
LLM_RERANKER | Model to use for reranking | "ms-marco-MiniLM-L-12-v2" |
LOCAL_LLM | Model to use when using local LLM via Ollama | "gemma2:9b" |
REMOTE_LLM | Model to use when using remote LLM fromm Grok | "llama-3.1-70b-versatile" |
LLM_TEMPERATURE | Model temperature parameter | "0.0" |
LLM_MAX_TOKENS | Max number of tokens for model | "8000" |
OLLAMA_BASE_URL | The base URL of Ollama. For example http://localhost:11434. When not set, then remote LLM is used via Grok. In this case, GROK_API_KEY must also be set | `` |
DEBUG | Enable debug logging | "true" |
CONVERSATION_MESSAGES_LIMIT | Max number of conversations with the uploaded files. Set to a negative number to not limits the number of conversations | "-1" |
GROQ_API_KEY | The API Key to use for Grok | "" |
Content type
Image
Digest
sha256:54a3f7800…
Size
791.7 MB
Last updated
over 1 year ago
docker pull startnow65/ragbase