Release - TestingCatalog (Page 3)

19 Dec 2025 · 1 min read

Google debuts FunctionGemma for on-device function calling

What's new? FunctionGemma is a function calling model from Gemma 3 270M for API actions; it runs on NVIDIA Jetson Nano and mobile phones with a 256k JSON vocabulary;

19 Dec 2025 · 1 min read

Mistral AI launches OCR 3 model for document parsing

What's new? Mistral OCR 3 wins 74% over its prior version in document text extraction; it powers Document AI Playground on Mistral AI Studio with an API at $2 per 1,000 pages;

17 Dec 2025 · 2 min read

Meta updates Edits app for creators with AI video effects and tools

What's new? Meta launched Edits, a mobile video app with AI effects and Reels integration; It offers 10-minute capture, watermark-free export and a frame-accurate timeline.

17 Dec 2025 · 1 min read

OpenAI opens ChatGPT to third-party apps after review process

OpenAI now lets developers submit third-party apps for listing in ChatGPT, opening a global path for tools to become usable directly inside ChatGPT.

17 Dec 2025 · 1 min read

xAI launches Grok Voice Agent API for real-time voice apps

xAI introduces the Grok Voice Agent API, offering developers real-time speech capabilities and configurable voice options for voice-first apps.

17 Dec 2025 · 2 min read

Google launches Gemini 3 Flash model with faster, low-cost results

What's new? Google launched Gemini 3 Flash for rapid processing and quality reasoning at low cost; it is now available via Google app, Search AI mode, and key APIs;

17 Dec 2025 · 2 min read

Google started rolling out Super GEMs from Opal inside Gemini

Google is rolling out the Super Gems feature in Gemini, merging Opal apps with the Gems Manager and introducing easier workflow-building tools.

16 Dec 2025 · 2 min read

OpenAI released GPT Image 1.5 on ChatGPT and APIs

OpenAI introduces an improved unified Images experience in ChatGPT, offering image creation, editing, and analysis within chat on web and mobile.

16 Dec 2025 · 1 min read

Meta launches SAM Audio for sound separation on playground

What's new? Meta unveiled SAM Audio, an AI model for sound segmentation; it supports text, visual, and span prompt methods on Segment Anything Playground;

15 Dec 2025 · 1 min read

NVIDIA launches Nemotron 3 Nano, Super and Ultra models

What's new? Nvidia unveiled Nemotron 3 models; Nano is live with 30b parameters and 1m-token context while Super and Ultra debut in 2026 with 4-bit NVFP4 on Blackwell;

15 Dec 2025 · 1 min read

Manus launches Max agent with new mobile and design tools

Manus version 1.6 introduces the Max agent with improved autonomous capabilities, mobile development tools, and a new Design View.

15 Dec 2025 · 2 min read

Google expands Gemini with NotebookLM integration

NotebookLM integration is rolling out to Gemini web, letting users attach notebooks as live data sources for conversations, supporting both free and paid accounts.

14 Dec 2025 · 2 min read

ICYMI: Opera launches Neon browser globally

Opera Neon is now available globally with a $19.90 monthly subscription, offering AI-powered agents, premium models, and experimental workflows.

12 Dec 2025 · 1 min read

World App adds encrypted chat, group payments, and virtual accounts

What's new? World App launches World Chat for encrypted texts, groups, media, payments and mini apps; virtual accounts, crypto support, and earn rewards roll out on iOS and Android.

12 Dec 2025 · 2 min read

OpenAI launches GPT-5.2 that scores 54% on ARC-AGI-2

OpenAI announces GPT-5.2, offering new models on ChatGPT and API, with advances in coding, spreadsheets, tool use, and expanded context windows.

12 Dec 2025 · 1 min read

Google releases Gemini Deep Research agent on APIs

What's new? Google released its Gemini Deep Research agent via the API with access to Gemini 3 Pro; it gathers online data and produces structured reports with JSON outputs;

12 Dec 2025 · 1 min read

Cursor launches visual editor for web app development in browser

What's new? Cursor has launched a visual editor in the Cursor Browser for React-based projects; it uses drag and drop layout, real time property adjustment and natural language prompts;

11 Dec 2025 · 1 min read

Google expands Gemini TTS with 24 languages, lifelike voices

What's new? Google announced Gemini 2.5 Flash and Gemini 2.5 Pro TTS preview models via the Gemini API in Google AI Studio; They offer lifelike speech synthesis with granular control over pace, tone, and 24-language support;