https://openmodelweights.com/news/uibk-embedding-v1-179m-unified-dense-colbert-retrieval-20261013/Open Model Weightsen2026-10-13T16:57:25ZUniversity of Innsbruck releases 179M UIBK-Embedding-v1 with dense and ColBERT retrieval in one index https://openmodelweights.com/news/alphaedge-sparseup-fr-v0-french-sparse-encoder-20261013/Open Model Weightsen2026-10-13T16:57:25ZAlphaEdge AI releases French Sparseup-fr-v0 retrieval weights after 55-hour training run https://openmodelweights.com/news/vllm-q4-2026-agent-serving-roadmap-20261013/Open Model Weightsen2026-10-13T11:31:02ZvLLM publishes Q4 roadmap for agentic serving, model support and reliability https://openmodelweights.com/news/lighton-colgrep-agent-2b-local-code-search-20261013/Open Model Weightsen2026-10-13T11:31:01ZLightOn releases local ColGrep-Agent-2B for coding-agent repository search https://openmodelweights.com/news/hunmin-397b-cua-open-weight-gui-agent-transfer-20261012/Open Model Weightsen2026-10-12T16:20:03ZHunmin-CUA releases 397B open weights for computer-use capability transfer https://openmodelweights.com/news/lightonocr-3-apache-2-document-layout-models-20261012/Open Model Weightsen2026-10-12T16:20:02ZLightOn releases Apache-2.0 LightOnOCR-3 models for text and document layout https://openmodelweights.com/news/tencent-youtu-parsing-omni-5b-eu-license-20261012/Open Model Weightsen2026-10-12T16:20:01ZTencent releases Youtu-Parsing-Omni weights with a specific EU license warning https://openmodelweights.com/news/ai2-gpu-fair-share-training-scheduler-results/Open Model Weightsen2026-10-11T17:05:03ZAi2 Reports Faster GPU Scheduling for Model Training https://openmodelweights.com/news/perplexity-pplx-embed-v2-late-open-multimodal-retrieval/Open Model Weightsen2026-10-11T17:05:02ZPerplexity Opens pplx-embed-v2-late Retrieval Models https://openmodelweights.com/news/cloudflare-clef-omni-open-multimodal-decision-weights/Open Model Weightsen2026-10-11T17:05:01ZCloudflare Releases Clef-omni Open Multimodal Weights