qwen2.5-7b-qa360-grpo

Qwen2.5-7B-Instruct fine-tuned with GRPO on the QA360 5,000-scenario dataset.

(risk_level, risk_justification, automation_candidate, affected_modules, functional_tests, negative_tests, security_tests, accessibility_tests, api_tests, affected_modules, functional_tests, negative_tests, security_tests, regression_scope). Max score: 1.0.

Usage

from transformers import pipeline
gen = pipeline('text-generation', model='ExecuteAutomation/qwen2.5-7b-qa360-grpo')
print(gen(prompt, max_new_tokens=1024)[0]['generated_text'])