Our fine-tuned Qwen models now match GPT-6 and Opus 5.5. And we didn't take a lab's word for it. We tested them on our own private benchmark, built from real work in my company that no model has ever seen. Whenever a local model passes, GPT-6 loses that job.