Alibaba Qwen · MoE Architecture
Qwen3.8-2.4T-A95B (open weights)
The open weights promised when Qwen3.8-Max launched last week. 2.4T total, 95B active, 92 layers built as 23 cycles of three Gated DeltaNet MoE blocks to one Gated Attention MoE block. Context is 262,144 native, extensible to 1,010,000. Scores 92.6 on GPQA Diamond, 86.6 on Terminal-Bench 2.1 and 67.7% on SWE-bench Pro. Thinking mode is mandatory.
Open the source ↗Similar Language Models releases
- Clover-35B-A3B-Preview
DiffuSpace · Diffusion MoE LLM · Week 45, 2026
- Groot-280B-A16B-261023
GrootTeam · Open reasoning MoE · Week 45, 2026
- Apex-35B-V1.0
Apex Intelligence · Research automation LLM · Week 45, 2026
- Ling-3.0-flash-Sante
inclusionAI (Ant Group) · Medical LLM · Week 44, 2026
- MemT-1.0-Nano
MemTensor · Memory LLM · Week 44, 2026
- Kimi K3.1 Pro
Moonshot AI · Open-Weight Frontier · Week 44, 2026
More from Alibaba Qwen
- Qwen-Image-2.1
Alibaba Qwen · Open Image Generation · Week 40, 2026
- Qwen3.8-Omni-Flash
Alibaba Qwen · Omni Modal Agent · Week 39, 2026
- Qwen3.8-LiveTranslate
Alibaba Qwen · Live Speech Translation · Week 39, 2026
- Qwen3.8-Max-0902
Alibaba Qwen · Qwen3.8 Max · Week 36, 2026
- Qwen3.8-Flash-Next
Alibaba Qwen · Qwen4 Preview MoE · Week 35, 2026
- Qwen3.8-27B
Alibaba Qwen · Reasoning · Week 33, 2026
Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.