Liquid AI's smallest model yet LFM2.5-230M beats models 4X its size at data extraction, can run 'anywhere'

LFM2.5-230M proves that while 3-billion-parameter models like VibeThinker are solving advanced calculus, a ...

New agentic memory framework uses 118K tokens per query. LangMem burns through 3.26M.

NUS researchers' MRAgent framework reduces LLM agent memory retrieval to 118K tokens per query — vs. 3.26M for LangMem — using step-by-step reasoning.

Thubrikar Aortic Valve partners with MAC’s MEDICAL Group and expands the TAVI-1 CE Mark trial of the Optimum TAVI System™

The Optimum TAVI system has been named “Top Aortic Valve Implantation System 2026” by the publication MED TECH Outlook. A detailed interview of Dr. Thubrikar, author of the 1991 textbook “The Aortic ...

Unseen Reality Debuts URXR One, Bringing Full Spatial Computing to 93-Gram Glasses

Debuting at AWE 2026, the company’s first product combines a wide 90-degree field of view, high-resolution Micro-OLED displays, and ultra-low-latency ...

OpenAI's updated GPT-5.5 Instant is better at shopping, complex constraints, and understanding user intent — and it's already in the API

OpenAI is moving away from models that require heavy hand-holding and toward systems that can better infer the user’s goal, ...

Benjamin Rogojan

VentureBeat delivers news, analysis, and insights on AI, data, and security—helping business leaders stay ahead in the rapidly evolving tech landscape.

Xiaomi's HarnessX rewrites its own AI scaffolding mid-task — and smaller models gain the most

Xiaomi's HarnessX autonomously rewrites AI agent harnesses mid-execution, delivering +14.5% avg performance gains — and +44% ...

Mistral launches OCR 4, turning document extraction into a full enterprise AI play

Mistral AI's OCR 4 delivers structured document intelligence with bounding boxes, confidence scores, and self-hosted ...

Alibaba's model never trained as an agent — and improved agent performance across seven benchmarks

Real environments can't inject edge cases on demand. Alibaba's Qwen-AgentWorld simulates them — and outperformed ...

OpenAI unveils first custom AI inference chip, Jalapeño, with Broadcom — and its development was sped-up with OpenAI's own models

The companies attributed this speed to a deep software-hardware co-development process that actively used OpenAI’s own models ...

Intuit will show off how it rebuilt its AI infrastructure to support fast and complex tasks at VB Transform 2026

Customer expectations have shifted from simple, fast conversational interactions to complex agentic AI-powered tasks that ...

Anthropic launches Claude Tag, replacing its Slack app with a persistent AI teammate that learns, monitors and works autonomously

Anthropic has launched Claude Tag, a persistent AI agent for Slack that lets enterprise teams delegate work, automate tasks, ...