A LoRA fine-tuned version of Llama-3.2-3B-Instruct for agricultural question answering and crop advisory generation. This is the 3B parameter variant — more capable than the 1B version for complex agronomic reasoning while remaining lightweight enough for edge deployment.

Fine-tuned using Unsloth for fast, memory-efficient training on a single GPU.

Also available: 1B variant (lighter, faster inference)

Model Details

Field	Details
Base model	`unsloth/Llama-3.2-3B-Instruct-bnb-4bit`
Fine-tuning method	LoRA (Low-Rank Adaptation) via PEFT
Trainer	SFTTrainer (TRL) + Unsloth
Precision	4-bit quantized (bitsandbytes)
Language	English
Domain	Agriculture / Crop Advisory
Developed by	shehryars715

Intended Use

✅ Direct Use

Answering crop-specific agricultural questions (pest control, irrigation, fertilization, disease management)
Generating advisories from remote sensing inputs (NDVI, EVI, crop growth stage)
Decision support for farmers, agronomists, and precision agriculture systems

✅ Downstream Use

Backend for agri-advisory chatbots and mobile apps
Integration into precision agriculture dashboards
RAG pipelines augmented with real-time field sensor data

❌ Out-of-Scope Use

Medical, legal, or financial advice
Non-agricultural domains
Autonomous high-stakes decisions without human expert review

Training Details

Training Data

Two datasets were mixed during fine-tuning:

General Agricultural Advisories — structured Q&A with fields: crop, topic, question, answer
Farm Sensor Data Advisories — precision agriculture records with fields: crop_type, stage, NDVI, EVI, and generated Advisory

Training Hyperparameters

Parameter	Value
Training regime	fp16 mixed precision
LoRA rank (r)	16
LoRA alpha	16
LoRA dropout	0
Target modules	q, k, v, o, gate, up, down proj
Batch size	2 (per device)
Gradient accumulation	4 steps (effective batch = 8)
Epochs	3
Learning rate	2e-4
Optimizer	AdamW 8-bit
LR scheduler	Linear
Max seq length	2048
Hardware	Kaggle (NVIDIA T4 / P100, single GPU)

shehryars715/
finetuned-llama-3.2-3b

🌾 Agricultural Advisory LLM — Llama-3.2-3B LoRA

Model Details

Intended Use

✅ Direct Use

✅ Downstream Use

❌ Out-of-Scope Use

Training Details

Training Data

Training Hyperparameters

How to Get Started

With Unsloth (recommended — fastest)

shehryars715/finetuned-llama-3.2-3b

🌾 Agricultural Advisory LLM — Llama-3.2-3B LoRA

Model Details

Intended Use

✅ Direct Use

✅ Downstream Use

❌ Out-of-Scope Use

Training Details

Training Data

Training Hyperparameters

How to Get Started

With Unsloth (recommended — fastest)

shehryars715/
finetuned-llama-3.2-3b