Qwen3.8 2.4T A95B

Qwen3.8-2.4T-A95B - First Qwen-Max-class open-weight model

Alibaba's Qwen3.8-2.4T-A95B is the first Qwen-Max-class model released as open weights. It is a sparse mixture-of-experts model, built on the Qwen3.5 architectural foundation and designed for coding, research, professional work, and long-horizon agentic tasks.

Hybrid linear and full attention

The defining architectural feature is a hybrid attention layout that alternates between Gated DeltaNet (linear attention) and Gated Attention (full attention). The 92-layer stack follows a repeating pattern of 23 blocks, each containing 3 Gated DeltaNet layers followed by 1 Gated Attention layer, all paired with MoE. Linear attention replaces the growing KV cache with a bounded recurrent state, keeping compute and memory bounded as context scales toward one million tokens.

Built for agentic completion

The model is designed to carry complex, multi-step tasks through to completion with stronger autonomous planning and better handling of environment feedback. Reasoning depth can be tuned per request - dialed up for multi-step reasoning or dialed down for high-throughput document processing. Day-0 support shipped for vLLM, SGLang, and TokenSpeed, with FP8 and BF16 checkpoints available at launch.