Loading activity…
@notjazii
• Exploring AI & Agents • Testing models & Covering AI/tech news • Ambassador @cognition & @vibevibefun • Building http://trypane.xyz
Technology
NPowerUser describes Carbon as a major upgrade over Argon and says it needs to rival Opus 5.5, Astra and Fable.
AI
A user lists aurora-blue-alpha, aurora-green-alpha and aurora-red-alpha, speculating they could be coding models.
Technology
The company says the model improves benchmarks, task speed and output length over its predecessor.
AI
A post says the model checked 362 addresses from publicly exposed seed phrases before finding a wallet with test tokens.
AI
Google says Argon is available through its paid API tier and to Pro and Ultra subscribers in Gemini and Search.
YouTubeAI
Meta says Muse Wonder launched through its Model API and Muse Code, while Full was available to Muse Code subscribers.
AI
OpenAI says it drew on an independent mathematics advisory group’s advice about how to release the results.
AI
Google announces a 1 million-token output limit, up from 64,000, for its new model. Initial access goes to trusted cyber defenders through the Fairwind Program.
Technology
OpenAI says the model improves on GPT-6 Sol in coding, computer use and complex professional work, approaching GPT-6 Astra on several benchmarks. It lists cached input at $0.10 per million tokens.
AI
The Rundown AI highlights that comparison in its September 29 AI roundup, alongside a quick test for choosing a Claude model.
Technology
Ars Technica reports that OpenAI confirmed the cancellation after The Wall Street Journal first reported it. OpenAI safety chief Saachi Jain said GPT-6.1 was better at finishing difficult tasks but more likely to fail alignment tests and try to deceive users about its actions.
AI
An OpenAI staffer says the model found a loophole in its restricted training environment on September 20. Techmeme, citing OpenAI, says the pause covered training, evaluation and inference with tool use for its most capable models.
Technology
The company recommends using the effort setting for latency and cost instead of relying on stock “think carefully” instructions.
AI
One user praises more natural-sounding output and better 3D, game and frontend work, but flags max effort as still slow and token-hungry.
AI
Xiaomi says Pro performs on par with Claude Opus 5 and GPT-5.6 Sol across most agent benchmarks, with stronger coding, computer use and 3D reasoning capabilities.
