Cainew

MiniMax M2.7 is a self-evolving AI model that helped build its own capabilities. It can create agent harnesses, collaborate via Agent Teams, and handle complex tasks like coding, debugging, and research. With strong SWE-Pro performance and reduced intervention time, it moves beyond static AI into systems that continuously learn, adapt, and execute complex work with minimal human input. Available via API and MiniMax Agent for builders pushing AI-native workflows.

Three new Kitten TTS (text-to-speech) models have been released, with the smallest being less than 25MB in size.

For teams building cloud systems, InfrOS designs and validates inherently optimized architectures that align to your priorities. It doesn’t just predict outcomes, it proves them through emulation before deployment - and helps you evolve infrastructure with control over time.

OctoClaw gives you AI specialists that actually execute business tasks - writing content, qualifying leads, replying to customers, and coordinating workflows across your tools.

Start a project on Netlify with just a prompt—no repo or local setup needed! Describe your app, pick an AI agent (Claude, Gemini, or Codex), and get a working, live production URL immediately. Iterate in place on real Netlify infrastructure with built-in forms and serverless functions. Traditional methods still work, but now you can go from prompt to real product instantly, without migrating later. Git is there when you're ready. Fastest way to try it: netlify.new

Comet is the first agentic AI browser built for mobile — summarize across tabs, chat with voice, organize your browsing, block distractions and let your AI assistant take action while you stay in control. The smartest way to browse on the go. Comet on iOS is now live in the Apple App Store.

Scouts for iOS is the best way to have AI agents research and monitor anything that matters to you, on the go.

Lucent is an AI that watches your session replays 24/7, and automatically detects issues your users are running into in real time. Trusted by leading engineering teams like Reducto, Browser Use, and Productlane.

Budibase introduces AI agents for operations teams. Handle requests, approvals, and workflows automatically - connected to your data and tools. Trigger from Slack/Teams/Discord, build apps when needed, and let agents do the work.

Doodles AI is a first-of-its-kind, artist-led AI platform under which sits Prism 1.0, a closed-loop model trained exclusively on Doodles IP. The model allows anyone to generate studio-grade images interpreted through Doodles iconic lens in seconds, without style theft or third-party IP infringement. For Doodles and their brand partners, it's an always-on UGC engine. Every generation is empowering your audience to create beautiful co-branded visuals at scale.

NVIDIA NemoClaw is an open source stack that simplifies running OpenClaw always-on assistants safely. It installs the NVIDIA OpenShell runtime, part of NVIDIA Agent Toolkit, a secure environment for running autonomous agents, with inference routed through NVIDIA cloud.

AI-powered scheduling agent that reads your emails, checks your calendar, and drafts perfect replies. Stop the back-and-forth: let AI handle your scheduling.

Meet GB1 — the private, planet‑friendly AI assistant from the UK. Built as an alternative to big tech AI, GB1 runs on 100% renewable UK energy, never trains on your chats, and keeps your data in the UK. Powered by Locai L1 Large, the first British foundational AI model, GB1 delivers frontier performance without trading privacy, principles or the planet. Create Spaces for projects, chat across web and mobile, and help shape the roadmap by voting on features. Available now on web & mobile

Meet Pluto Door — the macOS SSH client that makes plain terminals feel ancient. Connect to any server with terminal, file browser, code editor, and AI in one beautiful app. Debug configs, edit files, drag-and-drop uploads, and stay in flow. 100% local and secure — keys and passwords stay on your Mac in Keychain. Just $12. No subscriptions, no BS.

Machine Payments Protocol (MPP) is the open standard that lets AI agents pay for services programmatically.

How OpenAI uses chain-of-thought monitoring to study misalignment in internal coding agents—analyzing real-world deployments to detect risks and strengthen AI safety safeguards.

2% of ICML (International Conference on Machine Learning) papers were desk rejected because the authors used large language models (LLMs) in their reviews.

Researchers are investigating how many branches a CPU can predict to improve performance.

A survey of 81,000 people has revealed their desires and expectations from AI technology.

Accelerates Codex growth to power the next generation of Python developer tools