ArshansGithub

Migrating Prompts from Autoregressive to Diffusion LLMs

A practical guide based on migrating 18 production AI operations (~175 test cases) from GPT-4.1-mini to Mercury 2, a diffusion-based LLM. Every rule below was learned from a real failure and validated with automated tests.

---

How Diffusion LLMs Differ

Autoregressive models (GPT, Claude, Gemini) generate one token at a time, left to right. Each token sees everything before it. They follow instructions well because they process them sequentially while generating.

ArshansGithub

class EmailVerification:
    def __init__(self, email, password):
        self.server = IMAPClient('outlook.office365.com', port=993, use_uid=True)
        self.server.login(email, password)
        self.server.select_folder('INBOX')

    def amount(self):
        return self.server.select_folder('INBOX')[b'EXISTS']
        
    def check(self, target, match, methodType="text", delete=True):