by Qwen
Use Qwen/Qwen2.5-72B-Instruct-128K via the AIHubMix unified API — one interface for every major LLM.