Why agentic AI needs open data, and why synthetic data is how we scale it. Image: Nemotron Post-Training v3 Prompt Atlas More Than Model Weights Building AI agents is hard, because the real world does not behave like a benchmark. An agent that can’t recover from a broken API call, or a workflow it has…
November 19, 2024 — Posted by Jason Jabbour, Kai Kleinbard and Vijay Janapa Reddi (Harvard University)Everyone wants to do the modeling work, but no one wants to do the engineering.If ML developers are like astronauts exploring new frontiers, ML systems engineers are the rocket scientists designing and building the engines that take them there.Introduction”Everyone wants…
Listen to the article 3 min This audio is auto-generated. Please let us know if you have feedback. Dive Brief: Premiums for health plans on the Affordable Care Act exchanges are poised to spike again next year as policy upheaval and rising healthcare costs continue to roil the insurance marketplaces, according to a report released…
Data center builder Crusoe Inc. is reportedly in talks to raise a $3 billion funding round. Bloomberg on Thursday cited sources as saying that the deal could value the company at $30 billion. That’s about three times what Crusoe was worth last year. The report didn’t name the investors that may participate in the new…
Home » Blog » Google DeepMind’s Gemma 4: MoE, Efficiency Tricks, and Benchmarks Table of Contents Google DeepMind’s Gemma 4: MoE, Efficiency Tricks, and Benchmarks Gemma 4 Model Family Overview: E2B, E4B, 31B, and MoE 26B A4B Gemma 4 Capabilities: Reasoning, Multimodal AI, and Thinking Mode Gemma 4 Architecture Overview: Shared Transformer Design Principles Gemma…
xAI has released Grok 4.5. The model was trained on tens of thousands of Nvidia GB300 GPUs and targets coding, agentic tasks, and knowledge work. Benchmark results paint a mixed picture. On Terminal Bench 2.1, which tests complex command-line tasks, Grok 4.5 scores 83.3%, nearly matching GPT 5.5 (83.4%) and trailing Anthropic’s Fable 5 (84.3%) by just…
💡 Today, MiniMax released the MiniMax M3 model as open weights. Modular is proud to be a Day Zero launch partner. MiniMax M3 is the newest open-weights model that has been optimized for coding, agentic work, and native multimodality for MiniMax. A few things that make this a frontier model are: A 1M-token context window…
What You Should Know Cancer AI pioneer Hurone AI has entered into a strategic know-how agreement to license Mayo Clinic’s HOUSES (HOUsing-based index of Socioeconomic Status) technology. The collaboration natively embeds Mayo Clinic’s validated social determinants of health (SDoH) tool directly into Hurone’s governed cancer navigation ecosystem. The proprietary HOUSES metric calculates an individual’s socioeconomic…
RALEIGH, N.C. and ARMONK, N.Y., July 8, 2026 — IBM and Red Hat today announced the commercial launch of Lightwell, delivering automated vulnerability remediation at scale through two offerings: Lightwell Network and Lightwell Clearinghouse Premier. Available now, Lightwell Network gives enterprises access to a launch catalog of 6,500+ remediated, digitally signed, and certified application-layer dependencies…
some of the problems we face in implementing AI algorithms, we usually focus on the processors’ ability to handle them. But now, NVIDIA’s GPUs power the latest large language models, and companies compete to build faster AI accelerators. These new chips promise more computational power, more cores, and more operations per second. So, one would…