@testingcatalog

–

07 January 2026 9h47

Cursor release a dynamic context discovery for all models which reduced total tokens by 46.9%

7 January 2026

Gemini hits 20% AI chatbot traffic share

By

–

07 January 2026 9h16

Gemini surpassed 20% traffic share threshold among the overall traffic for AI chatbots. Grok is above 3% now as well

7 January 2026

AI Studio Adds Tools UI and Auto Function Response

By

–

06 January 2026 22h22

AI Studio got a new UI component for Tools, directly in the prompt bar. In addition, the "Automatic Function Response" option is in the works. "Gemini automatically generates a response for each function call"

NVIDIA CEO praises multi-model AI future

By

–

06 January 2026 16h51

NVIDIA CEO says that multi-model systems will be the future of AI.

Jensen Huang: “…@perplexity_ai was using multiple models at the same time, I thought it was completely genius.”

We've also seen Poetiq crushing ARC-AGI-2 with a system of models from different labs. pic.twitter.com/sg8tdBm3Hw
— 🚨 AI News | TestingCatalog (@testingcatalog) 6 janvier 2026

NVIDIA CEO says that multi-model systems will be the future of AI. Jensen Huang: “…
@perplexity_ai was using multiple models at the same time, I thought it was completely genius.” We've also seen Poetiq crushing ARC-AGI-2 with a system of models from different labs.

GPT-5.2-Codex-Max Testing Reports Surface

By

–

06 January 2026 16h14

OpenAI might be testing GPT-5.2-Codex-Max already with some users as there are multiple reports where Codex claims to be upgraded. Mine is still 5.1, what did you get? h/t @Haha58208745

Runway and NVIDIA demo video generation on Rubin

By

–

06 January 2026 15h44

Runway partnered with NVIDIA to be among the first to demonstrate video generation models on the NVIDIA Rubin platform. "Runway’s Gen-4.5, the world's top-rated video generation model, was ported from NVIDIA Hopper to Vera Rubin NVL72 in a single day"

COCOON: Decentralized AI Inference on TON

By

–

03 January 2026 21h52

Yes, according to their docs "COCOON is a decentralized AI inference platform on TON that securely connects GPU owners who provide compute with privacy-conscious applications that need to run AI models. For GPU Providers, it defines how suitable hardware can become part of a

3 January 2026

Telegram launches AI summaries with Cocoon

By

–

03 January 2026 21h51

Telegram released its first AI feature powered by Cocoon. New AI Summaries will become available on long-form posts and Instant View pages.

This feature is powered by open-source models running on Confidential Compute Open Network created by Telegram. pic.twitter.com/eKkgTYbrrh
— 🚨 AI News | TestingCatalog (@testingcatalog) 3 janvier 2026

Telegram released its first AI feature powered by Cocoon. New AI Summaries will become available on long-form posts and Instant View pages. This feature is powered by open-source models running on Confidential Compute Open Network created by Telegram.