AI Capex is Now Driven by Inference
markjeffrey · x · 2026-07-20
The core viewpoint shared is that the primary driver of AI capex is not training, but inference demand.\n\nThe author argues that while Chinese model training efficiency has proven to be highly competitive, this doesn't mean the model layer competition is over. What is truly exploding is the continuously growing inference demand. Whether users choose Claude or Kimi, the need for inference compute will trend upward long-term, and even hyperscale cloud providers might not be able to absorb all of it alone.
More from Infra
- The shadow GPU capacity market is formalizing, with Meta selling excess compute to outside buyers — DavidLinthicum · 2026-09-11
- Engram's random reads don't suit SSDs; CPU-memory over NVLink could serve all 72 GPUs — bookwormengr · 2026-09-11
- 80% of the DIY LLM inference hype posters have already quit — it's brutally hard systems work — abhijithneil · 2026-09-11
- Hugging Face's Ultra Scale Playbook: a free book on training LLMs on GPU clusters — mdancho84 · 2026-09-11
- Is inference latency becoming the biggest bottleneck for production AI agents? — Euphoric_Sea632 · 2026-09-11
- LLM Serving Metrics Thread: Why TPOT and Uptime Make or Break User Experience — abhijithneil · 2026-09-11