InclusionAI, released 2026-10-13. Hybrid MoE decoder, 560B total, 25B active, 1M context. Attention: 7:1 Kimi Delta Attention and gated MLA.
Sparse hybridAlibaba Cloud 🇨🇳 · February 16, 2026
Qwen3.5 (397B)
Mainline Qwen refresh that brings the Next-style hybrid attention into the flagship series.
Key detail
Turns the former Qwen3-Next side branch into the new core design with 512 experts and 17B active parameters.Total params
397B
Active / token
17B
Context
256K
Fact sheet
- Scale
- 397B total, 17B active (4.3% active)
- Context window
- 262,144 tokens
- Decoder type
- Sparse hybrid
- Attention
- 3:1 Gated DeltaNet and Gated Attention
- Layer mix
- 15 gated attention + 45 DeltaNet
- KV cache / token (bf16)
- 30 KiB · Low
- License
- Apache License 2.0
- Released
Efficiency
Parameters active per token
4.3%
17B of 397B weights run for each token.
AA Intelligence Index v4.3.2
Source21.4
- Scientific
- 10.3
Architecture concepts
Sources & resources
Use the ← and → keys to step through releases.
Similar architectures
Upstage, released 2026-07-22. Hybrid MoE decoder, 250B total, 15B active, 1M context. Attention: 3:1 Kimi Delta Attention and gated GQA with NoPE.
Alibaba Cloud, released 2026-08-26. Hybrid MoE decoder, 125B total, 6B active, 256K context. Attention: 3:1 Gated DeltaNet and Qwen Sparse Attention (QSA).
InclusionAI, released 2026-08-04. Hybrid MoE decoder, 124B total, 5.1B active, 256K context. Attention: 5:1 Kimi Delta Attention and gated MLA.