Imagine creating custom datasets and training AI models WITHOUT writing a single line of code. We did and made it a reality. @huggingface Synthetic Data Generator. Blog: Space: GitHub:

🔥 Bespoke curator: Synthetic Data Curation for Post-Training & Structured Data Extraction. Create synthetic data pipelines with easy!.- Retries and caching included.- inference via LiteLLM, vLLM, and popular batch APIs.- asynchronous operations. 🔗 URL:

🔥One > token > at > a > time < a < at < token < One🔥. token-explorer is a tool that lets you explore different possible paths that an LLM might sample!. - Arrow keys to navigate, pop and append tokens.- View the token probabilities and entropies. GitHub:

💥 SMASH and run models 5x faster, 5x cheaper . Pruna AI is the AI Optimization Engine for ML teams seeking to simplify scalable inference. Make sure to ⭐️ their GitHub: .TechCrunch: .Smashed Models on HF:

Agents aren't fully there yet, so we still rely on humans to create courses at Hugging Face! And with humans come slight imperfections. The @LangChainAI LangGraph framework unit will be released one week later on the 18th of March!. In the meantime, keep learning and creating!

Hook up AutoTrain, and you'll be able to go prompt to model!

🍽️ Let’s dissect the Synthetic Dataset Generator. 💬 Natural language prompt to data . 🦙 Ollama ensures secure local LLM inference. ✍🏼 Argilla’s data curation capabilities complete the workflow. 🔗 GitHub:

Private Synthetic Data Generation Made Easy: Out-of-the-Box with @Docker, @argilla_io & @ollama!. Synthetic data? All you need is Docker!. Thank you for the great contribution, @mcdaqc!. 🔗 Blog: 📺 YouTube:

🔗 Blog: 💪🏽 Great work @gretel_ai and team: Lipika R., Maarten Van Segbroeck, Ph.D., Dhruv Nathawani.

🔥 Dataset drop for SAFER alignment!. 🙋 Synthetic data containing safe/unsafe alignment pairs across various risk categories. 🔒 Helps mitigate malicious Use, System Risks, Information Hazards, Discrimination, and Societal Risks. 🔗 Dataset:

RT @mcdaqc: Privacy in AI is crucial, but how can we generate realistic training data without exposing sensitive info?. We explored a solut….

RT @dvilasuero: Weekend project:. Open-sourcing "Natural Science Reasoning", a smol reasoning dataset, the pipeline, and the code to build….

RT @nathanhabib1011: Deepseek skyrocketed 17,694% in just 12 weeks lmao. That’s what happens when you drop open models that rival the best….

RT @yifeiwang77: No More Tears with Matryoshka🪆 & Let's Embrace Sparsity 🚀🚀. With sparse autoencoders + sparse contrastive learning, we can….

@calebfahlgren great work by you and the team :).

🔥 Text2SQL, explore and share any data analysis!. 🤗 Hugging Face - Dataset Studio is an amazing new feature. 🚀 Start yourself:

The team has been working flat-out on this for a few weeks. Supported by @LoganMarkewich @seldo from @llama_index!. Who won? You decide!.

It is set to offer a great contrast with the smolagents unit by looking at. - What makes llama-index stand-out.- How the LlamaHub is used for integrations.- Creating QueryEngine components.- Using agents and tools.- Agentic and multi-agent workflows.

🥊 Epic Agent Framework Showdown! Available today!. 🔵 In the blue corner, as challenger, the experienced knowledge retriever: @llama_index. 🛑 In the red corner, the defender, weighing in with lightweight efficiency: @huggingface smolagents!. 🔗 URL:

🔥 Vicinity: SEVEN semantic search BACK-ENDS, ONE single INTERFACE!. 🫸 New release to push vector search to the Hub and work with any serialisable objects. 🧑‍🏫 KNN, HNSW, USEARCH, ANNOY, PYNNDESCENT, FAISS, and VOYAGER. 🔗 Library:

@Gradio is making AI accessible! ❤️.