Google unveils Gemma 4 12B.
Agentic reasoning, vision & audio
Runs locally on 16GB VRAM
Near-large-model performance
New unified multimodal architecture
Apache 2.0 open license
Powerful multimodal AI is becoming more accessible than ever.
Building apps just got easier.
OpenAI is making app creation more accessible with Sites in Codex.
With Sites in Codex, teams can turn ideas, plans, and workflows into interactive websites or apps that can be explored and shared with a URL.
MiniMax Introduced M3 — the first open-weights model to combine three frontier AI capabilities:
Frontier Coding & Agentic Performance
59.0% SWE-Bench Pro | 66.0% Terminal Bench 2.1 | 74.2% MCP Atlas , 1M-Token Context Window
Claude Opus 4.8 is here.
Sharper judgment
More honest progress updates
Longer independent task execution
Same pricing as Opus 4.7
The next generation of AI agents is becoming smarter and more autonomous.
Google introduces Google AI Threat Defense
A new AI-powered cybersecurity solution designed to continuously monitor and stop AI-driven threats before they impact businesses.
Real-time monitoring
Faster threat detection
AI-first cyber defense
Anthropic just made secure AI-assisted development more accessible.
A new security-guidance plugin for Claude Code is now available through the plugin marketplace, helping developers identify and fix vulnerabilities directly while writing code.
Capafy has introduced an AI Skills Marketplace a new infrastructure layer for monetizing AI-native workflows.
The platform supports integrations across Claude Code, Codex, and OpenClaw turning AI skills into scalable digital assets.
Google is releasing new agentic features for Pomelli, its AI tool for helping businesses create on-brand content.
Pomelli can now build a company’s “Business DNA” from scratch by pulling context from existing files and images.
Gemini 3.5 Flash is here — delivering faster inference, stronger reasoning, and next-gen agentic AI capabilities.
Running at 289 tokens/sec, it’s ~4x faster than many frontier models and optimized for Antigravity 2.0 multi-agent workflows — at just $1.50 per 1M input tokens.
Google AI Studio now supports:
• One-click Android app installation to devices
• AI-generated apps integrated with Google Workspace
• Rapid prototyping powered by Gemini 3.5 Flash
AI development workflows are moving closer to real-time software generation and deployment.
GOOGLE : A 24/7 GEMINI SPARK AI AGENT HAS BEEN ANNOUNCED!
> Comes with a dedicated virtual machine
> Supports MCPs and Connectors
> Powered by Gemini 3.5 and Antigravity harness
OpenAI launched a personal finance feature in ChatGPT for Pro users in the US.
You connect your bank accounts via Plaid, get a spending dashboard, and can ask GPT-5.5 questions grounded in your actual transaction data - balances, spending patterns, subscriptions, investments.
OpenAI launched GPT-Realtime-2 in the API — bringing GPT-5-level reasoning to real-time voice agents.
Alongside GPT-Realtime-Translate and GPT-Realtime-Whisper, developers can now build smarter, faster, and more conversational AI voice interfaces.
ChatGPT is now a direct add-on for Excel & Google Sheets.
GPT-5.5 brings advanced reasoning, automation, and natural language analytics directly into spreadsheet workflows.
Announced live at Code with Claude, Anthropic introduces “Dreaming” (research preview) — a step toward autonomous, goal-directed agent systems.
• Outcomes-based execution
• Multi-agent orchestration
• Webhooks in public beta
Anthropic has launched 10 pre-built AI agent templates for financial workflows such as pitchbooks, KYC, and month-end close.
These agents integrate with Excel, PowerPoint, Word, and Outlook, enabling cross-tool automation and orchestration of finance operations.
Voice cloning is now available on xAI Console in the US.
> Create a custom voice in less than 2 minutes or select from our library of 80+ voices across 28 languages to personalize your voice agents, audiobooks, video game characters, and more.
Manus AI announced Cloud Computer, an always-on machine that works for you 24/7!
Always-on agent as an infrastructure 🔥
> Users can host their own 24/7 Bots for Discord and Telegram via Cloud Computer
> Build a Persistent Knowledge Base
> Deploy and host any open-source tool
Gemini now generates Docs, Sheets, Slides & PDFs directly in chat.
No tool switching.
No manual formatting.
From prompt → production-ready output in one flow.
Already live for all users.