Tagged: mixture-of-experts

20 articles

Latest

Alibaba's Qwen3.8-Flash-Next: 125B params, 6B active, 1/9th the cost

Alibaba's Qwen3.8-Flash-Next hits 1/9th training cost of prior model

This Open-Source Engine Runs a 753B Model on a Single Workstation GPU

Mixtral 8x7B puts sparse Mixture of Experts models at the center of open AI

Mixtral 8x7B Beats Llama 2 70B and GPT-3.5 With an Apache 2.0 License

Alibaba's 2.4T param Qwen3.8-Max dethrones DeepSeek on Arena.AI, but at 14x the cost

Cursor drops MoK megakernel: 2.37x faster MoE training, but it demands a $2M rack

Inkling-Small beats its 975B teacher on reasoning, runs on a single GPU

AMD ships a 16B-param MoE model trained entirely on its own GPUs, but the license is research-only

Microsoft's new AI defense model hits 95.95% bug-finding rate but can't exploit a thing

Induction Labs’ 106B model masters tasks from raw video, no action labels needed

Laguna S 2.1 found a proof for a 1975 math problem — and it's an 8B model

Poolside's Laguna S 2.1 crushes DeepSeek on DeepSWE with one-sixth the active parameters

China's Moonshot Just Dropped a 2.8T Parameter Model That Trades Compute for Memory

Thinking Machines drops Inkling: a 975B-param open model that sees, hears, and reads

Soofi S: The 30B German moe model that spanks 70B rivals

OpenAI Ships Open-Weight Reasoning Models, First Since GPT-2

Transformers 4.49 quietly fixes the mess that made MoE models a deployment nightmare

Hugging Face just made GPT-OSS's best tricks part of every model

SegMoE packs 4 Stable Diffusion models into one pipeline — and prompt accuracy jumps