Meituan Trained a 1T+ LLM on Ascend — Why Doesn't Walmart Do the Same?
teortaxesTex · x · 2026-09-19
A discussion on whether large enterprises can train/evaluate frontier models. Respondent notes this works in China: Meituan ("Chinese DoorDash", 111K full-time employees) was the first to train a >1T parameter LLM on Huawei Ascend hardware, and JD and China Telecom also train serious LLMs. Why doesn't Walmart, with 2M+ employees, follow suit?
More from Companies & People
- UC Riverside opens cybersecurity postdoc role with salary up to $85,736 — pkqzy888 · 2026-09-19
- Exclusive: OpenAI's Sam Altman to brief UN Security Council next week — gaganghotra_ · 2026-09-19
- Perplexity now ships a new product or feature every 55 hours, down from 93 in 2025 — cameronstow · 2026-09-19
- Scale AI CEO Alexandr Wang says he personally writes all his launch tweets — alexandr_wang · 2026-09-19
- State of AI report to host launch meetups in SF, NYC and Paris — nathanbenaich · 2026-09-19
- Karpathy's year-old rant calling AI agents "slop" resurfaces as agent hype rolls on — steipete · 2026-09-19