AI Agents
AI Agent technology and applications
Articles
Google Research Releases ToolGrad: Answer-First Framework Hitsβ8
Google Research's ToolGrad flips tool-use data generation to answer-first, achieving a 99.8% pass rate and letting Gemma-3 match frontier models with just 500 s...
Redis LangCache: Managed Semantic Cache Cuts LLM API Costs by Upβ8
Redis LangCache is a managed semantic cache that matches prompts by meaning, cutting LLM API costs up to 90% and returning cache hits 15x faster.
OpenAI Launches Agents API in Public Beta: One API Call Nowβ7
OpenAI's Agents API is now in public beta, letting developers run Codex-powered agents with one API call via managed sandboxes, sessions, and tools.
NVIDIA Details BioNeMo Inference Runtime (BioIR): 2.90x Higherβ7
NVIDIA BioIR delivers 2.90x faster Boltz-2 folding and 58.5K residues per GPU-hour on 8xH100, scaling proteome-wide structure prediction in PyTorch.
DeepSeek-V4.1-Flash Launches with 1M Context, FP4 KV Cache, andβ9
DeepSeek-V4.1-Flash launches with 1M context, FP4 KV cache, and cross-layer attention reuseβcutting KV memory to 890 bytes per token for input-heavy agent workl...
LandingAI Launches Agentic Document Extraction Gen2 with DPT-3β7
LandingAI's Agentic Document Extraction Gen2 rebuilds document intelligence with DPT-3 Pro and Verity, using hierarchical tree modeling, character-based pricing...
Google Open-Sources Mantis: A Modular Skills Toolkit Empoweringβ8
Google releases Mantis, an open-source toolkit that helps coding agents find, reproduce, and patch software vulnerabilities with a modular, slash-command pipeli...
Gradium Launches Voice Design: Write a Prompt, Get a Brand Newβ8
Gradium Voice Design turns written descriptions into custom synthetic voices in secondsβno recording needed. Live on API and Studio with free tier access.
Meta Introduces Muse, a Personal AI Agent Running on a Secure,β7
Meta launches Muse, a proactive AI agent on a secure dedicated cloud VM, handling tasks like emails and negotiations with user approval.
NVIDIA Unveils CUDA Rust with cuda-oxide and cutile-rs forβ8
NVIDIA introduces CUDA Rust with cuda-oxide and cutile-rs, enabling compile-time-safe GPU kernels in Rust and boosting memory safety for AI development.
