A new wave of open-weight Mixture-of-Experts models like Ling, Qwen MoE, and Kimi K3 is combining huge parameter counts with low active parameters, enabling frontier-level quality on consumer hardware at a fraction of the cost.
A Hugging Face user has released a quantized Kimi K3 variant that strips multilingual support and applies IQ2-XXS via Unsloth, shrinking the footprint from 711GB to 478GB. Behind the technical move lies a clear lesson for on-premise evaluators: cutti...
A power user pushes a 32GB AMD Radeon AI Pro R9700 to its limits with INT4-quantized Qwen3.6 models. The 35B MoE achieves 262,144 token context and 52 tok/s at 100k depth, while the 27B leverages speculative decoding to sustain 59 tok/s past 50k. The...
Alibaba launched Qwen3.8-Max, a 2.4-trillion-parameter MoE model, while DeepSeek offers V4-Flash at $0.14 per million input tokens. Both release weights under open licenses, enabling on-premise deployment. Task cost hinges on token generation and int...
InclusionAI released Ling-3.0-flash, an open-weight MoE with 124 billion total parameters but only 5 billion active per token. Announced before the Kimi K3 and DeepSeek-V4-Flash wave, its sizing could carve a niche in on-premise deployment, where eff...
Three open-weight LLMs match human agreement on IMO-GradingBench at costs up to 100X lower than frontier models. The unanimous voting rule proves most stable, opening paths for local, sovereign, low-TCO evaluations.
Microsoft Research’s open-source Orchard framework leverages a shared Kubernetes environment to let compact models (3B active parameters) achieve near-frontier performance on software engineering, web navigation, and personal assistant tasks. The reu...
