📊 Natural Language to SQL

SLM Text-to-SQL

Translates natural language questions into valid SQL queries against custom database schemas offline.

🚀 Overview & Capabilities

Translates natural language questions into valid SQL queries against custom database schemas offline.

Key Features

Offline schema parsing and SQL translation
Dialect support: PostgreSQL, SQLite, MySQL
Guarantees valid syntax without leaking data to external servers

💻 Installation

Install the local package using pip:

Terminal

              $pip install slm-text-to-sql

🐙 Checkout from GitHub

Clone only this agent's folder from the monorepo using Git sparse-checkout — no need to download the full repository:

Option 1 — Sparse Checkout (Recommended)

Terminal — Git Sparse Checkout

              # 1. Create and enter a new directory
$ mkdir slm_text_to_sql && cd slm_text_to_sql

# 2. Initialise empty git repo and add remote
$ git init
$ git remote add origin https://github.com/t00114218-stack/SLMAgents.git

# 3. Enable sparse-checkout and set target folder
$ git sparse-checkout init --cone
$ git sparse-checkout set slm_text_to_sql

# 4. Pull only that agent's source
$ git pull origin main
            

Option 2 — Full Repository Clone

Terminal — Full Clone

              $ git clone https://github.com/t00114218-stack/SLMAgents.git
$ cd SLMAgents/slm_text_to_sql
            

💡 Tip: After checkout, install the package locally with pip install -e ./slm_text_to_sql to run in editable mode without publishing to PyPI.

⚙️ Configuration API

Constructor Parameters

Instantiate SLMTextToSQL with performance and runtime options:

Parameter	Type / Default	Description
`model_path`	str \| None	Path to local ONNX model. Default: None.
`temperature`	float \| 0.0	Sampling temperature. Default: 0.0.
`top_p`	float \| 0.9	Nucleus sampling threshold. Default: 0.9.
`max_tokens`	int \| 256	Max token output limit. Default: 256.
`n_threads`	int \| 4	CPU thread count. Default: 4.

Methods

Method Signature	Return Type	Description
`generate_sql(schema, query, ...)`	str	Generates valid SQL query string from natural language prompt.

Execution Parameters

Complete list of execution parameters accepted by the primary agent method:

Parameter	Type / Default	Description
`schema`	str	Database DDL creation schema string.
`query`	str	Natural language user question.
`system_prompt`	str \| None	SQL dialect prompt override. Default: None.
`user_input`	str \| None	Additional SQL constraints. Default: None.
`temperature`	float \| 0.0	Sampling temperature. Default: 0.0.
`top_p`	float \| 0.9	Nucleus sampling probability. Default: 0.9.
`max_tokens`	int \| 256	Maximum token limit. Default: 256.

Quick Start

Python Example

              from slm_text_to_sql import SLMTextToSQL

sql_agent = SLMTextToSQL(temperature=0.0, top_p=0.9, max_tokens=256)
query = sql_agent.generate_sql(
    schema="CREATE TABLE sales (id INT, amount DECIMAL, region TEXT);",
    query="What is total sales in West?",
    system_prompt="Dialect: PostgreSQL",
    user_input="Add WHERE amount > 0 condition",
    temperature=0.0,
    top_p=0.9,
    max_tokens=256
)
print(query)
            

🔍 Verified Output Logs

Diagnostic execution console output running locally on CPU:

Output Console

              → INPUT:
Query: What is total sales in West?

← OUTPUT:
"SELECT SUM(amount) FROM sales WHERE region = 'West' AND amount > 0;"
            