Alibaba (Qwen team) · Sparse MoE LLM
Qwen4-Max
Qwen4-Max is a proprietary sparse MoE language model from Alibaba's Qwen team, with a 2.4T-parameter backbone, 71B parameters active per token, a 1M-token context window and text, image and video input. It adds Qwen Sparse Attention, Gated Residual connections and 328B N-gram Embedding parameters. Alibaba reports first place on 4 of 16 charted benchmarks, but trailing on ProgramBench (31.5 vs 65.0 for Claude Opus 5.5) and Terminal-Bench 4.0 (48.5 vs 67.2 for Kimi K3.1). It is available only through hosted APIs at $1.50 input and $4.50 output per million tokens, with no open weights.
Open the source ↗Similar Language Models releases
- K2-Horizon
Institute of Foundation Models (K2 Team) · Open Model Fleet · Week 48, 2026
- LongCat-2.5
Meituan (LongCat team) · Agentic MoE LLM · Week 47, 2026
- Inkling 2 Small
Thinking Machines Lab · MoE agentic LLM · Week 46, 2026
- Ling-3.2-flash (Ling model series)
Ant Group · Multimodal MoE LLM · Week 46, 2026
- Kimi K3.1 Pro
Moonshot AI · Open agentic MoE · Week 46, 2026
- Hy4 preview
Tencent Hy Team · Open MoE LLM · Week 46, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.