Jacqkues commited on
Commit
ca9d4fb
·
verified ·
1 Parent(s): b24beb5

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +25 -1
README.md CHANGED
@@ -12,4 +12,28 @@ tags:
12
  short_description: Connect to more than 20 datasource and ask question
13
  ---
14
 
15
- Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
12
  short_description: Connect to more than 20 datasource and ask question
13
  ---
14
 
15
+
16
+ # Ibiza Server
17
+
18
+ Ibiza Server is a lightweight MCP server designed for **natural language question answering** over multiple SQL-compatible databases or files sources like csv or parquet. It empowers users to ask questions in plain English (or any supported language), then generates safe and optimized SQL queries using large language models and reranking techniques.
19
+
20
+ I have this for the Agent-MCP-Hackathon in the track : mcp-server-track.
21
+
22
+ A huge thanks to the **Modal** team for providing compute credits that enabled me to deploy two essential services for Ibiza:
23
+
24
+ - A reranking API powered by Qwen3-Reranker-4B, used to identify the most relevant tables based on the user’s question.
25
+
26
+ - A SQL generation service using Qwen3-14B with vLLM, which generates validated SQL queries and natural language explanations.
27
+
28
+ ## ✨ Key Features
29
+
30
+ - 🔌 **Multi-database compatibility**: Connects to MySQL, PostgreSQL, SQLite, Snowflake, and more via [Ibis](https://ibis-project.org/).
31
+ - 🧠 **AI-powered SQL generation** using **Qwen3-14B** via a custom VLLM inference server.
32
+ - 🛡️ **SQL validation and transpilation** using [SQLGlot](https://github.com/tobymao/sqlglot).
33
+ - 📊 **Table relevance reranking** with **Qwen3-Reranker-4B** via a dedicated reranking API.
34
+ - 🗣️ **Natural language answers** returned alongside query results.
35
+
36
+ ## Demo
37
+
38
+ - With an online **PostgreSQL** database : https://www.youtube.com/watch?v=CC9Z1wHNDPA
39
+ - With an online **CSV** dataset : https://www.youtube.com/watch?v=XiOmvVkuesg