Mistral AI has released Shieldstral 1.0 3B, an open-weights, policy-adaptive multimodal safety classifier that treats content moderation as a single yes/no question rather than a fixed taxonomy of harm categories. Most guardrail models bake their category list into the weights, so re-targeting one to a new deployment context means retraining — and the same content […]
Learn how the SAS Viya MCP Server enables AI assistants like Claude Cowork to orchestrate governed, auditable banking analytics workflows while keeping model execution, governance and oversight within SAS Viya. The post Building a governed banking workflow with SAS® Viya® MCP Server and Claude Cowork appeared first on SAS Blogs.
Building a governed banking workflow with SAS® Viya® MCP Server and Claude Cowork Read More »
Prime Intellect has open-sourced Prime Agent, a self-improving coding harness designed around two abstractions, the Recursive Language Model (RLM) and Continual Harness. Fixed tool schemas and context compaction force a model to work around its own scaffolding. Prime Agent replaces both with a persistent Python REPL and a rewritable harness. With Opus 5, it reports
NVIDIA has released Alpamayo 2 Super, a 34B-parameter vision-language-action (VLA) model for autonomous driving, under an open commercial license. The stated design target is the long-tail events: rare, multi-agent situations that conventional detection-and-prediction stacks handle poorly. The model pairs a 32B VLM backbone, built on NVIDIA Cosmos 3 Super Reasoner and post-trained with reinforcement learning,
Agents, MCP integrations, and LLM-powered applications are entering codebases faster than most security programs can track them. Mend.io’s new practitioner guide, ‘Securing AI agents, MCP servers & LLM apps: A practical framework’, targets that gap. It is organized around three moves: see what matters, fix what matters faster, protect AI in production and ships seven
How to Secure AI Agents, MCP Servers, and LLM Apps in Production Read More »
Thinking Machines Lab has released Inkling-Small, an open weights Mixture-of-Experts model with 276B total parameters and 12B active. That is about a quarter the size of Inkling, which carries 975B total and 41B active parameters. The model was trained on NVIDIA GB300 NVL72 systems. It reasons natively over text, images and audio. The context window
Agentic reinforcement learning research is constant algorithm modification. New estimators, new pipeline stages, new rollout schemes. In mainstream frameworks each change threads through layers of trainer, distributed backend, and rollout glue. That cost lands on the researcher at every iteration. Molt, from NVIDIA’s NeMo team, targets that cost directly. Its a PyTorch-native agentic RL framework
NVIDIA AI Releases Molt: A PyTorch-Native Agentic Reinforcement Learning Framework Read More »
In this tutorial, we implement an end-to-end streaming 3D reconstruction pipeline with LingBot-Map. We begin by configuring the input source, reconstruction settings, checkpoint selection, and output controls, then probe the available GPU and automatically tune frame limits, camera iterations, scale frames, and KV-cache parameters according to the detected VRAM. We install the repository and its
LingBot-Map Tutorial: GPU-Aware Inference and Point Cloud Export Read More »
In this tutorial, we build and execute a multi-agent workflow with Omnigent using a reliable, isolated Python environment created with uv. We configure a financial research lead agent that retrieves a live USD-to-EUR exchange rate from an external API, prepares a concise client-ready summary, and delegates its draft to a dedicated text-auditing sub-agent for clarity
Building a Policy-Governed Multi-Agent Financial Research Workflow with Omnigent Read More »
Tencent has released AngelSpec, an open-source, torch-native training framework for speculative-decoding draft models. The release covers both autoregressive multi-token prediction (MTP) and the block-parallel DFlash family. Most speculative-decoding work searches for one drafter that scores well on an averaged benchmark mixture. Real serving traffic does not look like that mixture. AngelSpec treats workload heterogeneity as
