Parameter	Value
Base Model	`unsloth/Qwen3.5-0.8B`
Method	Supervised Fine-Tuning (SFT) with LoRA
LoRA Rank	16
Dataset	thegovind/azure-architecture-vqa (1,678 train / 187 test)
Training Time	42.6 minutes on RTX 4090
Final Loss	0.6517
Steps	210 (1 epoch)
Hardware	1x NVIDIA RTX 4090 (24GB)

How to Use

python

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

base_model = AutoModelForCausalLM.from_pretrained(
    "unsloth/Qwen3.5-0.8B", 
    torch_dtype=torch.float16, 
    device_map="auto"
)
model = PeftModel.from_pretrained(base_model, "thegovind/azure-architect-qwen35-0.8b")
tokenizer = AutoTokenizer.from_pretrained("thegovind/azure-architect-qwen35-0.8b")

prompt = "What Azure service is best for global content delivery?"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
with torch.no_grad():
    output = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Pipeline

This is the first stage of a two-stage fine-tuning pipeline:

Base Model (Qwen 3.5 0.8B)
    → SFT (this model) — learns Azure knowledge
        → GRPO (thegovind/azure-architect-qwen35-0.8b-grpo) — learns structured reasoning

Related Models & Resources

Resource	Link
SFT LoRA (this)	thegovind/azure-architect-qwen35-0.8b
SFT Merged	thegovind/azure-architect-qwen35-0.8b-merged
GRPO LoRA	thegovind/azure-architect-qwen35-0.8b-grpo
GRPO Merged	thegovind/azure-architect-qwen35-0.8b-grpo-merged
Training Dataset	thegovind/azure-architecture-vqa
Benchmark	thegovind/azure-architecture-grpo-benchmark