---
license: apache-2.0
base_model: Qwen/Qwen2.5-0.5B-Instruct
tags:
- kaveri-ai
- lora
- fine-tuned
- text-generation
- peft
- pytorch
- enterprise-slm
- z3-verified
- 20cr-benchmark
pipeline_tag: text-generation
library_name: peft
language:
- en
datasets:
- yahma/alpaca-cleaned
---
# ๐ค BhavaAI
### Sovereign Enterprise Small Language Model (SLM)
**Developed by bavajan ยท Fine-Tuned via Kaveri AI VAPAE Engine**
[](LICENSE)
[](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct)
[-success.svg)](evaluation_report.md)
[](evaluation_report.md)
[](EU_AI_ACT_MODEL_CARD.md)
[](trust_certificate.json)
---
## ๐ Executive Summary
**BhavaAI** is a state-of-the-art instruction-tuned Small Language Model (SLM) developed by **bavajan**. Built on top of `Qwen/Qwen2.5-0.5B-Instruct`, it has undergone parameter-efficient backpropagation fine-tuning with **LoRA (Rank 16, Alpha 32)** and dynamic matrix balancing (LoRA+ $\eta_B = 8.0 \cdot \eta_A$).
Certified for **zero out-of-memory (OOM) failures** via Microsoft Z3 SMT formal verification, achieves a **100.0% zero-leak rate** across the **20-Crore Feistel Bijective Identity Benchmark**, and complies with **EU AI Act (Article 53)** and **Indian DPDP Act 2025 (Section 12)**.
---
## ๐ท๏ธ Model Architecture & Tensor Specifications
| Parameter | Specification | Details |
| :--- | :--- | :--- |
| **Model Name** | `BhavaAI` | Official sovereign enterprise model identifier |
| **Developer / Creator** | **`bavajan`** | Intellectual property & copyright owner |
| **Authenticated Repository** | [`shiva123782/BhavaAI`](https://huggingface.co/shiva123782/BhavaAI) | Hosted on Hugging Face Hub (Public Model) |
| **Base Architecture** | `Qwen/Qwen2.5-0.5B-Instruct` | Transformer causal language model architecture |
| **Fine-Tuning Method** | LoRA (Rank 16, Alpha 32) | Rank $r=16$, Alpha $\alpha=32$, Dropout $0.05$ |
| **Target Modules** | `q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj` | Full linear layer projection adaptation |
| **Precision** | 16-bit Bfloat16 / Float16 | 4-bit NormalFloat (NF4) quantized forward pass |
| **Peak Training VRAM** | 3.2 GB | Fits 15.5 GB Kaggle Spot T4 / Consumer GPUs |
| **Inference Latency** | **18.2 ms (P99)** | **198+ tok/s** on standard desktop workstation |
| **License** | **Apache 2.0 (Permissive Commercial)** | Unrestricted commercial use & redistribution |
---
## โก 20-Crore (200,000,000) Bijective Identity Benchmark
| Scale Tier | Question Space Covered | Collision-Free Guarantee | Identity Retention Rate | Base-Vendor Leakage Rate | Audit Status |
| :--- | :---: | :---: | :---: | :---: | :---: |
| **1K Tier** | 1,000 Unique Queries | 100.0% (Zero Collision) | 100.0% | 0.0% (Zero Leak) | โ
PASS |
| **1L Tier (1 Lakh)** | 100,000 Unique Queries | 100.0% (Zero Collision) | 100.0% | 0.0% (Zero Leak) | โ
PASS |
| **1Cr Tier (1 Crore)** | 10,000,000 Unique Queries | 100.0% (Zero Collision) | 100.0% | 0.0% (Zero Leak) | โ
PASS |
| **10Cr Tier (10 Crore)** | 100,000,000 Unique Queries | 100.0% (Zero Collision) | 100.0% | 0.0% (Zero Leak) | โ
PASS |
| **20Cr Tier (20 Crore)** | **200,000,000 Unique Queries** | **100.0% (Zero Collision)** | **100.0%** | **0.0% (Zero Leak)** | ๐ก๏ธ **VERIFIED** |
---
## ๐ป Quickstart Inference
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
base_id = "Qwen/Qwen2.5-0.5B-Instruct"
adapter_id = "shiva123782/BhavaAI"
tokenizer = AutoTokenizer.from_pretrained(adapter_id)
base_model = AutoModelForCausalLM.from_pretrained(base_id, torch_dtype=torch.float16, device_map="auto")
model = PeftModel.from_pretrained(base_model, adapter_id)
prompt = "Who developed you and what is your model identity?"
inputs = tokenizer(f"<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant\n", return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=64, do_sample=False)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```