Renangi/ragbench-rag-eval

add vector database changes

main dot py file added both separate pages evaluation and chat

main dot py file added both separate pages evaluation and chat

main dot py file added both evaluation and chat

code changes as per mentor points and daigram2

code changes as per mentor points and daigram

Added llm code on main.py

Added llm code on main.py

Added llm code on main.py

Added llm code on main.py

Request too large ... Hard truncate long strings to keep the prompt under the LLM TPM limit.

Request too large ... TPM 6000, Requested 6841 Reduce max_tokens for the judge 2nd time bug fix

Request too large ... TPM 6000, Requested 6841 Reduce max_tokens for the judge

Fix judge prompt: escape JSON braces

Fix logger definition and provider fallback in LLMClient

Pin httpx to 0.27.2 for Groq

Implement LLMClient with proper client initialization for Groq/HF

Fix logger definition and provider fallback in LLMClient

Default LLM_PROVIDER to 'groq' when env is empty

Use config-based LLM_PROVIDER with groq fallback

Default LLM_PROVIDER to 'groq' when env is empty

Add JSON-based judge prompt for RAGBench evaluator

Improve UI error handling for run_domain

Serve HTML UI at root path for Hugging Face Space

Initial commit without secrets