Open sourceHugging Face2 days agoDaisyRose-27B: Pushing the Frontier of Post-Training EfficiencyPrimary sourcecaught 10 h 6 min after publication
Open sourceHugging Face2 days agoWhy your fine-tuning job takes longer than the FLOPs sayPrimary sourcecaught 11 h 7 min after publication
Open sourceHugging Face2 days agoTime-Series Models in the Streamhouse: A New Frontier and Use CasesPrimary sourcecaught 15 h 1 min after publication
Open sourceHugging FaceNov 13Build a Robot Pick-and-Place Simulation with SimReady AssetsPrimary sourcecaught 12 h 0 min after publication
Open sourceHugging FaceNov 13XYBench: Helping with the problem behind the questionPrimary sourcecaught 12 h 58 min after publication
Open sourceHugging FaceNov 12LLMON: An LLM-Native Markup Language, now in Granite 4.1 3B and 8BPrimary sourcecaught 10 h 53 min after publication
Open sourceHugging FaceNov 9Nemotron-Personas-Finance-Japan: A Synthetic Dataset Capturing Japan's Demographics and Financial ContextPrimary sourcecaught 14 h 55 min after publication
Open sourceHugging FaceNov 9Your Experts Have Partners. There's HOPE for Keeping Them TogetherPrimary sourcecaught 17 h 26 min after publication
Open sourceHugging FaceNov 6ToyBlocks: A controlled world for robot learning and evaluationPrimary sourcecaught 16 h 0 min after publication
Open sourceHugging FaceNov 5New NVIDIA Cosmos 3 Open Models for Interactive World SimulationPrimary sourcecaught 10 h 26 min after publication
Open sourceHugging FaceNov 4Dharma Field Extractor: Any Document In, Structured Fields in JSON OutPrimary sourcecaught 12 h 11 min after publication
Open sourceHugging FaceNov 3StateBench: Testing how language models track state changes over long documentsPrimary sourcecaught 12 h 46 min after publication
Open sourceHugging FaceOct 29Granite Switch 4.1.1 — Update models like you update softwarePrimary sourcecaught 16 h 24 min after publication
Open sourceHugging FaceOct 28NVIDIA Nemotron 3.5 Transcribe and Granary v2: Intelligent Multilingual Transcription You Can Prompt and AdaptPrimary sourcecaught 15 h 3 min after publication
Open sourceHugging FaceOct 27Build specialized multimodal models with LFM3Primary sourcecaught 12 h 44 min after publication
Open sourceHugging FaceOct 22OlmoEarth v1.3: State-of-the-art embeddings for the whole planetPrimary sourcecaught 13 h 49 min after publication
Open sourceHugging FaceOct 22Ensemble forecasting with IBM Granite time-series modelsPrimary sourcecaught 22 h 9 min after publication
Open sourceHugging FaceOct 21NVIDIA Kumo Relational Outperforms Supervised Machine Learning on Relational Data, No Training RequiredPrimary sourcecaught 11 h 34 min after publication
Open sourceHugging FaceOct 21Profiling TorchNeuron (Part 1): System-level profiling on AWS Trainium.Primary sourcecaught 13 h 56 min after publication
Open sourceHugging FaceOct 21Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and FasterPrimary sourcecaught 15 h 3 min after publication
Open sourceHugging FaceOct 20NVIDIA Cosmos3-H-Dreams: Bringing Clinical Surgery to Life Through Generative SimulationPrimary sourcecaught 11 h 43 min after publication
Open sourceHugging FaceOct 20FairStress: A deeper evaluation of fairness in LLM decision-makingPrimary sourcecaught 13 h 52 min after publication
Open sourceHugging FaceOct 2Open-sourcing AstaBrief, the fast report-generation model in AstaAstaBrief, a fast report-generation model, has been open-sourced.Primary source