metadata

base_model: unsloth/Llama-3.2-3B-Instruct
tags:
  - text-generation
  - mongodb
  - query-generation
  - transformers
  - unsloth
  - llama
  - trl
  - gguf
  - quantized
license: apache-2.0
language:
  - en
datasets:
  - skshmjn/mongo_prompt_query
pipeline_tag: text-generation
library_name: transformers

MongoDB Query Generator - Llama-3.2-3B (Fine-tuned)

Developed by: skshmjn
License: apache-2.0
Finetuned from model: unsloth/Llama-3.2-3B-Instruct
Dataset Used: skshmjn/mongodb-chat-query
Supports: Transformers & GGUF (for fast inference on CPU/GPU)

🚀 Model Overview

This model is designed to generate MongoDB queries from natural language prompts. It supports:

Basic CRUD operations: find, insert, update, delete
Aggregation Pipelines: $group, $match, $lookup, $sort, etc.
Indexing & Performance Queries
Nested Queries & Joins ($lookup)

Trained using Unsloth for efficient fine-tuning and GGUF quantization for fast inference.

📌 Example Usage (Transformers)

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "skshmjn/Llama-3.2-3B-Mongo-Instruct"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)
schema = {} # Pass your mongodb schema here, leave empty for generic queries. Sample available in hugging face's repository

prompt = "Here is mongodb schema {schema} and Find all employees older than 30 in the 'employees' collection."
inputs = tokenizer(prompt, return_tensors="pt")

output = model.generate(**inputs, max_length=100)
query = tokenizer.decode(output[0], skip_special_tokens=True)

print(query)