AI Announcements - TestingCatalog AI News

1 Aug 2026 · 2 min read

xAI adds character references and 1080p to Imagine Video 1.5

Imagine Video 1.5 adds up to seven image references, voice consistency, prompt-only generation, and native 1080p for Grok users.

1 Aug 2026 · 2 min read

Thinking Machines launched open-weight Inkling-Small

Inkling-Small is now available with open weights, Tinker fine-tuning and multimodal Playground chat, using 12B active parameters and up to 1M tokens.

31 Jul 2026 · 1 min read

Gemini for macOS adds new "Speak to Window" feature

New Gemini voice tools are rolling out globally in English on macOS, with Fn-key dictation and optional reasoning based on on-screen context.

29 Jul 2026 · 2 min read

SpaceXAI launches Grok Voice Think Fast 2.0 on Agent Builder

Grok Voice Think Fast 2.0 is available at $0.08 per audio minute, with grok-voice-latest switching to the new model on August 5.

26 Jul 2026 · 2 min read

Anthropic launched Claude Opus 5 across all platforms

Claude Opus 5 is now the default model on Claude Max, the strongest on Claude Pro, and available through the API with optional faster mode.

26 Jul 2026 · 2 min read

ICYMI: Black Forest Labs opens FLUX 3 Video early access

FLUX 3 Video is now in early access, generating up to 20-second videos with native audio from text, images, keyframes or reference clips.

26 Jul 2026 · 2 min read

Microsoft previews MAI-Image-2.5-Pro and MAI-Voice-2-Flash

Microsoft has opened public previews of MAI-Image-2.5-Pro and MAI-Voice-2-Flash in Foundry, adding high-fidelity imaging and faster, cheaper speech.

25 Jul 2026 · 2 min read

ICYMI: Cursor launches Router to cut costs for coding models

Cursor Router is now available to Teams and Enterprise users across desktop, web, iOS, CLI, and SDK, with three modes for routing coding requests.

22 Jul 2026 · 2 min read

Google launches Gemini 3.6 Flash and Gemini 3.5 Flash Lite

What's new? Gemini 3.6 flash cuts token use for coding and data, and gemini 3.5 flash-lite runs at 350 tps; gemini 3.5 flash cyber targets vulnerability detection in pilot;

17 Jul 2026 · 2 min read

OpenAI unveils GPT-Red to boost AI safety with internal red-teaming

OpenAI introduces GPT‑Red, a specialized internal system for uncovering vulnerabilities in AI models using advanced red-teaming and self-play learning.

17 Jul 2026 · 2 min read

Perplexity launches SPACE runtime for AI agent tasks

Perplexity unveils SPACE, a new sandboxed runtime that powers all Computer sessions and enables persistent AI agent operations at scale.

15 Jul 2026 · 2 min read

Thinking Machines debuts open-weight Inkling AI model

Thinking Machines Lab launches Inkling, its open-weight Mixture-of-Experts AI model, offering broad customization and multi-modal support.

11 Jul 2026 · 2 min read

Cursor 3.11 adds side chats and agent transcript search

Cursor 3.11 adds Side Chats for parallel agent conversations, full transcript search, advanced agent controls, and improved project selection.

9 Jul 2026 · 3 min read

OpenAI launches ChatGPT Work for Pro, Enterprise, and Edu plans

OpenAI introduces ChatGPT Work, allowing users to execute projects, manage files, and automate workflows across apps and web tools on all platforms.

9 Jul 2026 · 3 min read

OpenAI launches GPT-5.6 Sol, Terra, and Luna on apps and API

OpenAI launches GPT-5.6 with Sol, Terra, and Luna models for ChatGPT, Work, Codex, and API, offering higher intelligence and faster artifact creation.

9 Jul 2026 · 2 min read

Anthropic adds usage reflection dashboard to Claude for all users

Anthropic introduces a reflection dashboard for Claude, letting users review activity trends, task types, and set quiet hours or break nudges via Settings.

9 Jul 2026 · 3 min read

Meta debuts Muse Spark 1.1 model and opens API for developers

Meta introduces Muse Spark 1.1 with a public API preview, offering advanced AI agent performance, an expanded context window, and improved multimodal workflows.

9 Jul 2026 · 1 min read

Mistral unveils Robostral model for robot navigation with RGB camera

What's new? Robostral Navigate is an 8-billion parameter model for embodied navigation that uses a single RGB camera to guide diverse robots; it was trained in simulation using 400k trajectories, tree-based token masking, and online reinforcement learning to boost navigation;