Moonshot AI releases Kimi K3, a 2.8T-parameter open model claiming frontier-class performance
“Open Frontier Intelligence”
MoE
9 tracked signals on MoE.
MoE architecture is a major trend in efficient AI model training.
Alibaba releases open weights for 2.4T-parameter Qwen3.8-Max, its largest open-weight model
Moonshot AI's Kimi K3 is the first open-weight model in the 3 trillion parameter class
“Kimi K3, a 2.8 trillion parameter Mixture of Experts (MoE) model that represents the first open-weight system to reach the 3 trillion parameter class”
Moonshot AI's Kimi K3 2.8T MoE beats Opus 4.8, claiming best open-weights model title
“This is more than a model drop; it is a fairly complete recipe for large-scale agentic post-training and serving.”
Hugging Face releases OLMo-core 3, open-source scalable training infrastructure for large MoE models
NVIDIA Nemotron 3.5 Lightning brings 4x agentic throughput to SageMaker JumpStart on a single GPU
NVIDIA launches Nemotron 3.5 Lightning, a 30B MoE model optimized for agentic execution layers
“Long-running AI agents spend most of their time on high-volume execution: tool calls, result validation, and subagent delegation.”
AWS achieves 40% throughput gain for MoE reinforcement learning using EKS, EFA, and DeepEP