Open-weight releases are increasingly centered on Mixture-of-Experts architectures with high total parameter counts, low active parameters and very long context windows. Chinese open models and new entrants continue to reduce the gap with closed US models.
XHToken has published Spark-X2.6-4B, an evolution of Spark-X2.5-4B. Same architecture, gains from improved post-training in coding, agents, long context and multilingual tasks. Three adjustable reasoning levels balance quality, latency and inference ...
Ornith-1.5-122B extends the self-improvement loop of Ornith-1.0: instead of optimizing only scaffold and rollout, it jointly generates tasks, scaffolds, and solutions through reinforcement learning. GGUF weights lower the barrier for local use, but t...
Reflection has released Beam, an open LLM with a 501-billion total and 23-billion active parameter MoE architecture. It supports a 1 million token context and BF16, FP8, and NVFP4 precision. The focus is token-efficient reasoning: about one third of ...
The model card for Tencent Hy-Memory-30B-A3B sparks debate over a slippery claim: an architecture with 3 billion active parameters out of 30 billion total would beat Opus 5 on an unspecified memory test. The issue is not the benchmark but the structu...
Meta has opened the weights of Muse Spark Open, but the point is not just scale: Helion, a PyTorch-native DSL for portable kernels, and TorchTitan support for fine-tuning shift the discussion to hardware control, TCO, and self-hosting. A second- and ...
Meta releases the weights for Muse Spark Open, a model with 816B total and 42B active parameters and a 1M-token context window. The announcement combines Helion and TorchTitan: portable kernels across NVIDIA, AMD, and Intel GPUs, plus PyTorch-native ...
inclusionAI's Ling-3.1-flash appears on Hugging Face with 560 billion total parameters, 25 billion active parameters per token, and a 1-million-token context window. The mixture-of-experts configuration reduces per-token compute, but shifts the focus...
An open-source leaderboard aggregates 22,113 benchmark results and shows China's open-weight models about seven months behind US closed models. The methodology separates 14,436 independent evaluations from 7,677 lab-reported scores. For on-prem deplo...
