Zyora-DEV-32B

Code Generation & Security Scanning Model

Try the API • Documentation • Website

Model Description

Zyora-DEV-32B is a fine-tuned large language model specialized for:

• Code Generation: Generate high-quality code in Python, JavaScript, Go, Rust, Java, C++, and more
• Security Scanning: Detect vulnerabilities including SQL injection, XSS, command injection, path traversal
• CWE Classification: Identify and classify vulnerabilities using Common Weakness Enumeration (CWE) IDs
• Auto-Remediation: Suggest fixes for detected security issues

Model Details

Property	Value
Base Model	Qwen2.5-Coder-32B-Instruct
Parameters	32.5B
Context Length	32,768 tokens
Fine-tuning Method	LoRA (merged)
Training Data	Curated code + security datasets
License	Zyora Community License

API Access

Zyora-DEV-32B is available via our OpenAI-compatible API:

```bash curl -X POST https://app.zyoralabs.com/v1/chat/completions
-H "Content-Type: application/json"
-H "Authorization: Bearer YOUR_API_KEY"
-d '{ "model": "Zyora-DEV-32B", "messages": [ {"role": "user", "content": "Write a Python function to merge two sorted lists"} ], "max_tokens": 512 }' ```

Pilot Program: Register at app.zyoralabs.com for 100,000 tokens/month free.

Usage with Transformers

```python from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "zyoralabs/Zyora-DEV-32B"

tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForCausalLM.from_pretrained( model_name, torch_dtype="auto", device_map="auto" )

messages = [ {"role": "user", "content": "Write a secure file upload handler in Python"} ]

text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True) inputs = tokenizer([text], return_tensors="pt").to(model.device)

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

License

This model is released under the Zyora Community License. See LICENSE for full terms.

Contact

• Website: zyoralabs.com
• API Portal: app.zyoralabs.com
• Email: support@zyoralabs.com