KV Cache Size Calculator
zephyr_z9 · x · 2026-07-18
Introduces a KV Cache Size Calculator tool designed to estimate the cache footprint of various models based on token length, sequence count, KV precision, and indexer precision.
Examples for DeepSeek V4 Pro and GLM-5.2 are provided: under 1 million tokens, single sequence, FP8/INT8 KV precision, and FP4/INT4 indexer precision, their total cache sizes are approximately 4.14 GiB and 43.09 GiB, respectively. The page also details the formulas and breakdowns, showing its capability to estimate inference KV/indexer costs across different architectures.
More from Infra
- DeepSpeed shows AutoEP/AutoTP/AutoSP parallelism gains, hosts office hours — PyTorch · 2026-10-27
- Cloudflare Durable Objects validated as top architecture for AI agent WebSocket servers — tristanbob · 2026-10-27
- Gary Marcus: With Q3 data added, Anthropic's compute-efficiency growth shows deceleration — GaryMarcus · 2026-10-27
- SpaceX locks in $40B GPU financing deal with Apollo, the largest ever committed — TechCrunch AI · 2026-10-27
- Inference is just Kubernetes with a multi-trillion-dollar market, and the communities overlap — matt_slotnick · 2026-10-27
- vLLM eyes Rust migration as inference engines race to rewrite the stack — AccBalanced · 2026-10-27