8 results for “Mixture of Experts”
HUMAIN unveiled humain-m3, a 428B-parameter Arabic model built on MiniMax-M3, at LEAP Riyadh on September 3. The research preview via HUMAIN Node gives developers early access to a model that claims the highest average score across seven Arabic benchmarks.
Alibaba's Qwen team released Qwen3.8-Flash-Next, a 125B-parameter multimodal MoE model with 6B activated parameters and 51B N-gram embeddings. The open-weight preview of Qwen4's architecture cuts training cost to one-ninth of its predecessor while leading coding and office-task benchmarks.
Moonshot AI has launched Kimi K3, a 2.8-trillion-parameter model with native multimodal capabilities and a 1-million-token context window. Open-model releases have mostly clustered well below the trillion-parameter class.
NVIDIA validated Alibaba's Qwen3.8-Flash-Next 176B MoE model on its GB300 NVL72 rack, achieving over 16K tokens per second per GPU. The hybrid GDN/QSA architecture keeps memory and compute bounded for million-token agentic coding workloads.
NVIDIA published Day-0 FP8 inference results for Alibaba's 2.4T-parameter Qwen3.8-2.4T-A95B on the GB300 NVL72 rack, hitting over 4,000 tokens per second per GPU. The result demonstrates that rack-scale NVLink can absorb MoE all-to-all traffic at frontier model scale.
NVIDIA released Nemotron 3.5 Lightning, a 30B MoE model with 3B active parameters built for high-volume agent execution. The open model cuts inference cost for always-on agents by handling routine tool calls while frontier models focus on planning.
Tencent open-sourced Hy4 preview, a 770B-parameter MoE model with 49B activated weights and a 1M-token context window. The Apache 2.0 release puts a frontier-class model within reach of self-hosted inference stacks.
Moonshot AI has released Kimi K3, a 2.8-trillion-parameter open-weight model that Moonshot positions as its largest public release to date. The strategic story is not just parameter count.