Qwen3.6 open-weight duo live (APIYI-hosted, no GPU rental)

Qwen3.6 series adds two open-weight models — qwen3.6-27b and qwen3.6-35b-a3b — hosted on APIYI’s official relay. Pay-as-you-go Chat flat pricing: 27b $0.42/$2.52, 35b-a3b $0.26/$1.56 per 1M tokens. Customers skip GPU rental and inference-framework ops.

2026/4/28 00:29 (UTC+8) · New Model · Alibaba

🚀 Qwen3.6 open-weight duo live (APIYI-hosted) — qwen3.6-27b (27B dense, coding rivals 397B-class) and qwen3.6-35b-a3b (same lineage as closed-source Flash MoE) open-weight models are now on APIYI: pay-as-you-go Chat flat pricing, 27b $0.42/$2.52, 35b-a3b $0.26/$1.56 per 1M tokens. Skip the GPU rental and inference-framework ops.

📖 Details: /en/news/qwen-3-6-opensource-launch

---

← Back to Live Updates · 📚 Monthly archive