How to use from
vLLM
Install from pip and serve model
# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "minktn/Qwen2.5-3B-Instruct-Text2SQL-Tuned"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "minktn/Qwen2.5-3B-Instruct-Text2SQL-Tuned",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'
Use Docker
docker model run hf.co/minktn/Qwen2.5-3B-Instruct-Text2SQL-Tuned:Q4_K_M
Quick Links

Qwen2.5-3B-Instruct-Text2SQL-Tuned

This model is a specialized Text-to-SQL adaptation inheriting directly from the base architecture Qwen/Qwen2.5-3B-Instruct. It is fine-tuned to translate complex natural language queries into executable, standardized SQLite syntax given relational schema definitions (DDL).


📌 Heritage & Model Overview

  • Base Architecture: Qwen/Qwen2.5-3B-Instruct
  • Domain Specialization: Text-to-SQL (Spider-style multi-domain benchmarks)
  • Primary Task: Schema-grounded query generation
  • Available Artifacts:
    • final_model/: Full-precision PyTorch / SafeTensors weights & tokenizers (for evaluation & fine-tuning pipelines).
    • qwen2.5-3b-instruct.Q4_K_M.gguf: 4-bit medium quantized artifact optimized for lightweight edge and local CPU inference.
Downloads last month
2
GGUF
Model size
3B params
Architecture
qwen2
Hardware compatibility
Log In to add your hardware

4-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for minktn/Qwen2.5-3B-Instruct-Text2SQL-Tuned

Base model

Qwen/Qwen2.5-3B
Quantized
(345)
this model