Dynamo prices GPU, RAM, and disk cache reuse differently
Abhishekcur · x · 2026-07-27
Cache reuse is treated as a discount
This part explains how Dynamo accounts for memory locality:
- If the prompt cache is still on the GPU, reuse gets the full discount.
- If it has to be copied from CPU memory, the discount is smaller.
- If it lives on disk, the discount is smaller still.
- These weights are explicit in the math, and they can be tuned down to zero if a deployment differs.
The key idea is that cache hit value depends on where the cache physically sits.
Related event: How NVIDIA Dynamo Prices LLM GPU Routing Instead of Hard KV Rules(14 posts)→
More from Infra
- Dedicated Hardware for Running AI Agents at Scale Arrives — cyrilzakka · 2026-09-23
- Ternary Bonsai 2 27B: 5.9GB weights retain ~95% of full-precision reasoning — cephaloform · 2026-09-23
- Qwen 27B runs 24hr unattended on one RTX5090, builds full Postgres-SpringBoot-React spreadsheet app — anglepoiselife · 2026-09-23
- OpenRoboto Shift launches: decentralized egocentric video data network for robot brains — markjeffrey · 2026-09-23
- Engineer describes designing digital circuits that recycle most of their energy — MikePFrank · 2026-09-23
- Cloudflare CTO Dane Knecht makes TIME's 2026 executives list as AI crawlers hit 52% of traffic — dinasaur_404 · 2026-09-23