📰 AI News — Latest AI Updates
Page 787 News
NVIDIA Jetson Memory Optimization: Running Larger AI Models at the Edge
As open-source generative AI models expand from data centers to edge devices, NVIDIA introduces memo...
NVIDIA Megatron Integrates Higher-Order Optimizers to Accelerate LLM Training
NVIDIA has integrated higher-order optimization algorithms such as Shampoo into its Megatron framewo...
End-to-End FP8 Precision Accelerates Reinforcement Learning Training
As large language models advance from text generation to complex reasoning, the computational cost o...
Three AI Agents Write 600,000 Lines of Code to Win Kaggle Competition
In March 2026, three large language model agents automatically generated over 600,000 lines of code ...
NVIDIA Launches RTX PRO 4500 Server Edition and vGPU 20
NVIDIA has unveiled the Blackwell architecture-based RTX PRO 4500 Server Edition GPU and vGPU 20 sof...
NVIDIA Introduces Universal Sparse Tensor to Simplify Deep Learning Development
NVIDIA has introduced Universal Sparse Tensor (UST) in its nvmath-python library, decoupling tensor ...
DeepSeek V4 Officially Launches Alongside NVIDIA Blackwell Support
DeepSeek releases its fourth-generation flagship models V4-Pro and V4-Flash, while NVIDIA provides G...
Microsoft Launches CORPGEN Benchmark to Drive AI Agents Toward Real-World Office Scenarios
Microsoft Research has released the CORPGEN benchmark framework designed to evaluate AI agents' abil...
Microsoft Launches PlugMem: Turning AI Memory from a Burden into an Asset
Microsoft Research has released the PlugMem framework, designed to transform the cluttered interacti...
Microsoft Releases Phi-4-reasoning-vision Multimodal Reasoning Model
Microsoft has officially released the 15-billion-parameter open-source multimodal reasoning model Ph...
Microsoft Launches AgentRx Framework for Systematic AI Agent Debugging
Microsoft Research has released the AgentRx framework, providing systematic debugging capabilities f...
Microsoft Launches AutoAdapt: An Automated Domain Adaptation Framework for Large Language Models
Microsoft Research has released the AutoAdapt framework, designed to tackle the domain adaptation ch...