Introducing the Private Hub: A New Way to Build With Machine Learning
Hugging Face launched 'Private Hub' offering dedicated, secure spaces for enterprises to host models and datasets with granular access controls.
Hugging Face Blog·3 Aug 2022
Use this view to inspect the underlying evidence corpus. For ranked developments, decision posture and interpretation, use Signals.
Raw feed or Signals?
Raw feed is chronological evidence. Signals ranks and interprets material change.
Hugging Face launched 'Private Hub' offering dedicated, secure spaces for enterprises to host models and datasets with granular access controls.
Hugging Face provided comments on the U.S. National AI Research Resource (NAIRR) Interim Report, advocating for open-source AI.
OpenAI research details efficient training of 'fill-in-the-middle' (FIM) language models, improving code generation and contextual completion.
OpenAI's Frontier Lab released a hazard analysis framework for LLM-based code synthesis, focusing on security and reliability risks.
Hugging Face blog details deploying TensorFlow Vision models via TF Serving, showcasing interoperability in model serving infrastructure.
Hugging Face released BLOOM, a 176B parameter multilingual open-access language model trained on 46 natural languages and 13 programming languages.
OpenAI trained an AI model using Video PreTraining on unlabeled human Minecraft videos to perform complex crafting tasks via native human interface.
Hugging Face Optimum now facilitates converting Transformer models to ONNX for optimized inference, targeting improved latency and throughput.
Intel and Hugging Face partnered to integrate Intel's AI hardware into Hugging Face's platform, aiming to optimize model training and inference.
Hugging Face published its third 'Director of Machine Learning Insights' report, focusing on AI adoption trends and challenges within the finance sector.
OpenAI research shows models writing critiques help humans identify more flaws in summaries, with larger models excelling at self-critique.
Enterprise AI leader Eugene Yan discusses applying software design patterns to ML code and system architecture for better maintainability and scalability.
OpenAI published a blog on the engineering challenges and techniques for training large neural networks at scale.
Hugging Face blog post explains the core components and mechanisms of diffusion models, a foundational generative AI architecture.
OpenAI, Cohere, and AI21 Labs published preliminary best practices for LLM deployment, aiming to standardize operational guidelines.
Graphcore and Hugging Face partnered to offer IPU-optimized transformer models, claiming performance benefits on Graphcore hardware.
OpenAI claims Codex powers 70 applications via API, implying broader adoption of code generation models for diverse use cases.
Hugging Face featured Sasha Luccioni, a notable figure in machine learning, in their 'Machine Learning Experts' series. No specific technical or deployment details provided.
Hugging Face launched a fellowship program for AI/ML researchers focused on open-source contributions, providing resources and mentorship.
Hugging Face published a blog post discussing machine learning insights for SaaS, focusing on operational metrics and value.
Hugging Face raised $100M in new funding, signaling continued investment in open-source AI platforms and model development.
Hugging Face now officially supports fastai, integrating fastai models and datasets into the Hugging Face Hub for broader access and sharing.
OpenAI announced executive role changes to its leadership team, reflecting recent progress and focusing on future milestones.
Hugging Face is hiring a Director of Machine Learning Insights for an 'Enterprise AI' focus, signaling an intent to deepen enterprise engagement.
Hugging Face launched a dedicated platform, "Hugging Face for Education," offering free access to compute and resources for academic and educational purposes.
Hugging Face blog post discusses applying machine learning to improve customer service, a common enterprise AI use case.
OpenAI blog post discusses Goodhart's Law in the context of optimizing AI objectives that are difficult or costly to measure, an internal challenge.
Hugging Face's blog features Lewis Tunstall, a machine learning expert, likely discussing advancements relevant to enterprise AI.
Habana Labs and Hugging Face are collaborating to optimize transformer model training on Habana Gaudi AI accelerators, targeting lower cost training.
Hugging Face introduced Decision Transformers, a model type for offline reinforcement learning, now available on their platform.