KV cache

Knowledge graph concept · also referred to as KV fetches, per-conversation state

Mentioned in

• Write-up of Reiner Pope's Lecture: How GPT, Claude, and Gemini Are Actually Trained and Served post · 2026-07-14

Related concepts

• AI inference pricing 1
• Inference batching 1
• LLM inference 1
• Memory bandwidth 1
• Mixture-of-experts model 1
• Roofline model 1
• Speculative decoding 1