Stories by gmays

Gemini API File Search is now multimodal

Powering the Inference Era: Inside the DigitalOcean AI-Native Cloud

Maryland Power Bills: $1.6B Data Center Subsidy

You Are Not Immune to Mode Collapse

Hugging Face's Clem Delangue: Stop Comparing Engines to Cars

Consumer AI's ARPU Problem

The context window has been shattered: Subquadratic debuts a 12M token window

Deepsec: The security harness for finding vulnerabilities

vLLM Routing and KV

Reduce friction and latency for long-running jobs with Webhooks in Gemini API

How to Scale Your Model

Meta plans advanced 'agentic' AI assistant for users

Model-Harness-Fit

Hallucinations Undermine Trust; Metacognition Is a Way Forward

Apple plans to make iOS 27 a Choose Your Own Adventure of AI models

Meta will use AI to analyze height and bone structure to identify underage users

End-to-End Autoregressive Image Generation with 1D Semantic Tokenizer

How LLM Inference Works

Replit's Amjad Masad on Cursor deal, Apple fight, and why he'd rather not sell

Coding plan comparisons based on actual usage

1 2 3 4 5 6 7 8 9 10