RakshakAI v3 (रक्षक AI)

Security-specialized fine-tune of Qwen2.5-Coder-7B-Instruct for vulnerability detection, CWE classification, and secure fix generation.

Fine-tuned with QLoRA on 80K curated CWE samples across 13 programming languages. Produces structured 9-field security reports.

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Muneerali199/rakshak-cwe-v3", device_map="auto")
tokenizer = AutoTokenizer.from_pretrained("Muneerali199/rakshak-cwe-v3")

Limitations

4-bit QLoRA adapter (requires base Qwen2.5-Coder-7B-Instruct)
2048 token context window
Best for Python, C, C++ (weaker on Ruby, Swift, Kotlin)

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for Muneerali199/rakshak-cwe-v3

Base model

Qwen/Qwen2.5-7B

Finetuned

Qwen/Qwen2.5-Coder-7B

Finetuned

Qwen/Qwen2.5-Coder-7B-Instruct

Finetuned

(448)

this model