RAG with one repo. If you want to load a new repos, call prepare_retriever(repo_url_or_path) first.

Custom implementation of Conversation to fix the list assignment index out of range error

A component wrapper for the Dashscope (Alibaba Cloud) API client. Dashscope provides access to Alibaba Cloud's Qwen and other models through an O

Manages the creation, loading, transformation, and persistence of LocalDB instances.

Model for LLM model configuration

Model for the entire model configuration

Model for LLM provider configuration

Model for the data to be stored in the wiki cache.

Model for requesting a chat completion.

Batch embedder specifically designed for DashScope API

LiteLLM OpenAI-compatible client. LiteLLM exposes an OpenAI-compatible API surface, so we can reuse almost all OpenAIClient behavior whi

Simple conversation management with a list of dialog turns.

Process documents for Ollama embeddings by processing one document at a time. Adalflow Ollama Client does not support batch embedding, so we

Test data pipeline functions that use embedders.

Test embedder configuration system.

Test embedder selection via environment variables.

Model for requesting a chat completion.

r""" A user-facing component that orchestrates an embedder model via the DashScope model client and output processors. Args: model_cl

Component that converts document sequences to embedding vector sequences, specifically optimized for DashScope API

Custom exception for when Ollama model is not found

Test individual embedder clients.

Test the embedder factory function.

Comprehensive tests for the _extract_repo_name_from_url method

Test the specific issues identified in the codebase.

Test RAG class integration with different embedders.

Model for the request body when saving wiki cache.

Model for requesting a wiki export.

Model for a wiki page.

Model for the wiki sections.

Model for the overall wiki structure.