Tagged: MoE

9 articles

Latest

Z.ai's GLM-5.3-Flash hits near-Opus coding scores at one-tenth the price

Alibaba's Qwen3.8-Max hits 86.6 on Terminal-Bench, open weights drop next week

Unsloth crushes MoE training with 7.3x speedup, Axolotl hits 1.45x on Qwen3.5

Alibaba drops 2.4T Qwen3.8 preview with no active params and no license yet

Kimi K3 hits near-frontier performance, but the era of dirt-cheap Chinese AI is over

NVIDIA's NeMo AutoModel hits 3.7x training speedup with a single line of code

MoE, Tiny Models, and Domestic Chips: China's Open-Source AI Playbook After DeepSeek

OpenAI drops GPT-OSS: a 120B open-source MoE model that fits on a single 80GB GPU

Meta drops Llama 4: 10M context window and a bold bet on NoPE layers