https://openmodelweights.com/news/ai2-astabrief-8b-open-weight-scientific-reports/Open Model Weightsen2026-10-02T15:00:00ZAi2 releases AstaBrief 8B for fast, cited scientific report generation https://openmodelweights.com/news/llama-cpp-adds-decision-models-system-one-endpoint/Open Model Weightsen2026-10-02T14:30:00Zllama.cpp adds decision-model inference through a new System One endpoint https://openmodelweights.com/news/olmo-core-3-open-moe-training-stack/Open Model Weightsen2026-10-02T13:40:00ZOlmo-core 3 opens a training stack designed for trillion-scale mixture-of-experts models https://openmodelweights.com/news/nvidia-kumo-tabular-open-foundation-model/Open Model Weightsen2026-10-02T12:50:00ZNVIDIA Kumo Tabular brings open foundation models to tabular prediction https://openmodelweights.com/news/holo4-open-agent-models-27b-35b/Open Model Weightsen2026-10-02T12:10:00ZHolo4 ships 27B and 35B-A3B agent models with open weights and trajectories https://openmodelweights.com/news/liquid-ai-lfm25-vl-dspark-speculative-decoding/Open Model Weightsen2026-10-02T11:30:00ZLiquid AI releases a 280M draft model to accelerate LFM2.5-VL-3B https://openmodelweights.com/news/flux-3-action-open-weight-world-action-model/Open Model Weightsen2026-10-02T10:55:00ZBlack Forest Labs releases FLUX 3 Action, a 7B open-weight world action model https://openmodelweights.com/news/nvidia-nemotron-3-diarization-open-weight-100m/Open Model Weightsen2026-10-02T10:20:00ZNVIDIA releases a 100M open-weight Nemotron 3 model for speaker diarization https://openmodelweights.com/news/transformers-runs-gguf-llama-cpp-quants/Open Model Weightsen2026-10-02T09:40:00ZHugging Face Transformers adds efficient GGUF inference using llama.cpp kernels https://openmodelweights.com/news/nvidia-agrees-to-acquire-hugging-face-open-platform-pledge/Open Model Weightsen2026-10-02T09:00:00ZNVIDIA agrees to acquire Hugging Face and says the platform will remain open