A publicly shared recipe applies Group Relative Policy Optimization via TRL to LiquidAI's LFM2.5-350M, raising its IFStruct score from 22.6% to 29.7% after 100 steps on roughly…
Jun Kim, the creator and maintainer of oMLX, has joined Hugging Face effective September 2026. The move aims to stabilize oMLX development, accelerate contributions to Apple's…
Transformers can now load and run GGUF files directly on Apple Silicon using ggml kernels, delivering near-llama.cpp performance while keeping the familiar Python and PyTorch…
The UK AI Security Institute is publishing verified benchmark results for frontier models using EvalEval's Every Eval Ever schema and Evaluation Cards, advancing reproducible…
Hugging Face’s summer ecosystem report separates attention from observed use. For model selection, repository popularity is a starting point, not a verdict.
A July research preprint examines changes in decision behavior after abliteration. Its narrow experiment raises a wider question about model modifications.
A downloadable checkpoint answers one question. Before using or distributing it, check the exact release, its terms and the provenance of any derivative.