NVIDIA LoRA & QLoRA Recipe Generator

# Auto-generated by NEXUS AI Fine-Tuning Cockpit
# Optimized for RTX 4090 24GB with Unsloth + FlashAttention-2
import torch
from unsloth import FastLanguageModel

max_seq_length = 2048
dtype = None # Auto detection
load_in_4bit = True # 4-bit QLoRA cuts VRAM by 75%

model, tokenizer = FastLanguageModel.from_pretrained(
    model_name = "meta-llama/Meta-Llama-3.1-8B",
    max_seq_length = max_seq_length,
    dtype = dtype,
    load_in_4bit = load_in_4bit,
)

# Apply LoRA adapters
model = FastLanguageModel.get_peft_model(
    model,
    r = 16,
    target_modules = ["q_proj", "k_proj", "v_proj", "o_proj", "gate_proj", "up_proj", "down_proj"],
    lora_alpha = 32,
    lora_dropout = 0,
    bias = "none",
    use_gradient_checkpointing = "unsloth",
    random_state = 3407,
)

from trl import SFTTrainer
from transformers import TrainingArguments
from datasets import load_dataset

dataset = load_dataset("timdettmers/openassistant-guanaco", split = "train")

trainer = SFTTrainer(
    model = model,
    tokenizer = tokenizer,
    train_dataset = dataset,
    dataset_text_field = "text",
    max_seq_length = max_seq_length,
    args = TrainingArguments(
        per_device_train_batch_size = 2,
        gradient_accumulation_steps = 4,
        warmup_steps = 10,
        max_steps = 100,
        learning_rate = 2e-4,
        fp16 = not torch.cuda.is_bf16_supported(),
        bf16 = torch.cuda.is_bf16_supported(),
        logging_steps = 1,
        output_dir = "nexus_lora_output",
    ),
)

trainer_stats = trainer.train()

# Save final adapters ready for NEXUS Private Vault deployment
model.save_pretrained_merged("nexus_final_model", tokenizer, save_method = "merged_16bit")
print("Training Complete! Ready to push to NEXUS Private Vault.")