---
license: apache-2.0
library_name: transformers
pipeline_tag: text-generation
tags:
- quantumindssi
- sovereign-ai
- edge-computing
- post-quantum-cryptography
- quantum-cryptanalysis
- nist-pqc
- vulnerability-detection
- ml-kem
- ml-dsa
- slh-dsa
- 01_quantum_resistant_crypto_analyzer
- finetuned
- lora
inference: true
---

# Quantum-Resistant Cryptographic Protocol Analyzer

A fine-tuned Small Language Model (SLM) that analyzes cryptographic protocol implementations and identifies quantum-vulnerable patterns, attack vectors, and NIST-aligned mitigations.

## Model Details

| Attribute | Value |
|-----------|-------|
| **Developer** | QuantumIndSSI Ltd |
| **Base Model** | ./base_model |
| **Architecture** | Transformer decoder (causal LM) |
| **Fine-tuning Method** | LoRA (Low-Rank Adaptation) |
| **LoRA Rank** | 16 |
| **LoRA Alpha** | 32 |
| **License** | apache-2.0 |

## Intended Use

- Automated quantum vulnerability scanning of protocol implementations (TLS, SSH, VPN, etc.)
- Security audit assistance for classical-to-PQC migration planning
- Developer education on quantum cryptanalysis risks
- Edge deployment on Victron and other constrained hardware

## Training Data

- 10,500+ synthetic cryptographic protocol analyses
- Protocols: TLS/SSL, SSH, IPsec, WireGuard, WPA, S/MIME, OpenPGP, DNSSEC, Kerberos, Bitcoin, Ethereum, gRPC, MQTT, Bluetooth
- Vulnerability types: Shor-vulnerable, Grover-amplified, HNDL, downgrade attacks, weak randomness, deprecated protocols, transition gaps
- Attack vectors: Shor factoring, Shor DLP, Grover search, quantum collision finding, HNDL passive collection, quantum MITM

## Evaluation Results

| Metric | Target | Score |
|--------|--------|-------|
| Perplexity | < 10.0 | TBD |
| Vulnerability Detection Rate | > 85% | TBD |
| Attack Vector Recognition | > 80% | TBD |
| Edge Latency (CPU) | < 1000ms | TBD |
| Memory Footprint | < 4GB | TBD |

## Usage

```python
model_id = "quantumindssi/01_quantum_resistant_crypto_analyzer"
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained(model_id)
tokenizer = AutoTokenizer.from_pretrained(model_id)

prompt = """Analyze the following TLS 1.2 implementation for quantum-vulnerable patterns:
```python
context = ssl.SSLContext(ssl.PROTOCOL_TLS_CLIENT)
context.set_ciphers('RSA-AES256-GCM-SHA384')
```
Identify the vulnerability type, quantum attack vector, and recommend mitigations."""

inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

## Limitations

- Not a substitute for certified security consultants or formal verification
- Synthetic training data may not capture all real-world edge cases
- English only
- Analysis is heuristic; false positives/negatives are possible

## Hardware Requirements

| Target | RAM | Notes |
|--------|-----|-------|
| Cloud GPU | 4GB | FP16 inference |
| Workstation | 3GB | INT8 quantized |
| Victron Edge | 2-3GB | INT8/INT4 quantized, CPU |

## Citation

```bibtex
@misc{01_quantum_resistant_crypto_analyzer,
  title={Quantum-Resistant Cryptographic Protocol Analyzer},
  author={QuantumIndSSI Ltd},
  year={2026},
  publisher={Hugging Face},
  howpublished={\url{https://huggingface.co/quantumindssi/01_quantum_resistant_crypto_analyzer}}
}
```

## Contact

- GitHub: https://github.com/QuantumindSSI
- HuggingFace: https://huggingface.co/quantumindssi
- Email: contact@quantumindssi.com