Loading activity…
@philhchen
Building something new. Previously research @openai @GoogleDeepMind @scale_AI @Stanford
AI
Core Automation describes two sites with different GPUs and parallelism schemes connected by a bidirectional 25 Gbps link.
AI
Claude says its proactive app will brief users on connected apps and seek approval for specified sensitive actions.
XAI
A user compares that potential shift to writing and the printing press, which spread intellectual output widely.
AI
A post contrasts CPUs' few fast cores with GPUs' many threads and lower clock speeds as metaphors for work.
AI
A post argues that direct pixel-space training has an advantage because even good simulators are costly and lossy.
AI
OpenAI says it drew on an independent mathematics advisory group’s advice about how to release the results.
AI
A post puts a GB300 NVL72 rack at roughly 1,580 kg and $5 million, or $3,165 per kilogram, then projects a much higher value per kilogram for future NVL72-class racks.
AI
One post argues that high model costs clash with a basic constraint: most people still cannot pay for software.
AI
A user asks whether others have used image inputs with Qwen3.5 35B-A3B on Tinker.
AI
One user calls both impressive but says they miss too many important details when given autonomy.
AI
Updated model targets coding, reasoning, and long-running agentic tasks.
