Loading activity…
@chetaslua
AI researcher & reporter | I test new models on launch day, build working demos, and post prompts + results | Code: http://github.com/ChetasLua
AI
A post claims an internal model scored 0.875 on BenchCAD 2.0, versus GPT-6 Astra's 0.594 at maximum.
AI
Motion says its Fable 5.5 workflow represents everything as editable code and exports to After Effects or Premiere Pro.
Technology
Google plans to start with a subset of cyber defenders, then expand access through its Fairwind Program.
AI
A post says Carbon scored 88.2% on SEC-bench Pro and is rolling out first to cybersecurity specialists.
AI
DeepSeek says the model is live on its website, app and API, and optimized for DeepSeekHarness and Tencent CodeBuddy.
AI
A user says it used multi-agent visual critique loops in Antigravity to sculpt a multi-building temple complex.
XTechnology
Reuters reports the deal, disclosed in Anthropic’s IPO filing, deepens ties spanning compute supply, equipment leasing and financing.
AI
The user presents what they call its first question: a request for advice based on smells, taste, warmth and light.
