live
- your daily tech & design digest, one email a day -
NAVER commits to gigawatt-scale sovereign AI infrastructure [Xtra!]South Korea's largest tech firm is betting big on domestic AI compute independence.
DeepSeek V4 Pro claims precision win over GPT-5.5 Pro [Xtra!]Another open-weight-adjacent Chinese lab claims to beat a flagship US frontier model on a key metric.
Xiaomi's MiMo Ultraspeed claims 15x speed advantage [Xtra!]Inference speed is becoming as competitive a battleground as raw capability.
Google ships Gemma 4 QAT models for on-device efficiency [Xtra!]Quantization-aware training keeps pushing capable models onto laptops and phones.
OpenAI reportedly prepping biggest ChatGPT overhaul since launch [Xtra!]The chat-first interface paradigm may be on its way out at OpenAI.
LangChain: give your agent its own computer [Xtra!]Sandboxed, persistent compute per agent is becoming standard architecture, not a nice-to-have.