Open Source
133 articles in this category (Page 6 of 6)
Microsoft Releases Agent Lightning: A Reinforcement Learning Framework for Optimizing AI Agents
Microsoft introduces Agent Lightning, an open-source framework that enables reinforcement learning (RL)-based training of large language models (LLMs) for AI agents without requiring changes to existing agent stacks.
Liquid AI Releases LFM2-ColBERT-350M: A Compact Late Interaction Model for Multilingual Cross-Lingual Retrieval
Liquid AI introduces LFM2-ColBERT-350M, a 350M-parameter late interaction retriever optimized for multilingual and cross-lingual search, offering high accuracy and fast inference speeds.
Sentence Transformers Joins Hugging Face as Community-Driven Open-Source Project
Sentence Transformers, a popular open-source library for generating sentence embeddings, has transitioned to Hugging Face. The project will remain community-driven and open-source, benefiting from Hugging Face's infrastructure and continued development.
Hugging Face AI Sheets Adds Vision Capabilities for Image-Based Data Analysis
Hugging Face releases a significant update to AI Sheets, introducing vision support to extract data from images, generate visuals from text, and edit images directly within a spreadsheet environment, powered by open-source AI models.
IBM Advances Open-Source AI with vLLM, torch.compile, and Spyre Accelerator Integration
IBM is significantly contributing to the open-source AI ecosystem by enhancing vLLM with hardware-agnostic kernels, achieving efficient LLM training with torchtitan, and integrating its Spyre AI accelerator for improved inference.
ALTK: Open-Source Toolkit Boosts Agent Reliability and Robustness
IBM Research introduces ALTK, an open-source toolkit to enhance the reliability and robustness of AI agents powered by large language models. ALTK provides modular components addressing various lifecycle stages, integrating with tools like ContextForge MCP Gateway and Langflow.
Leading the Way in Open Earth Observation AI
IBM Research, in collaboration with NASA and ESA, is advancing Earth observation AI through open-source models, compression techniques, and real-world applications, setting new standards for geospatial intelligence.