# gauravapiscean/agentic-kv-cache

> A prefix-cache simulator for LLM serving that replays real Claude Code and Mooncake traces to test whether published policies beat LRU. The author finds LRU is harder to beat than the papers claim.

- Magnitude: 0.1 out of 10 — Quiet
- Stars: 23 total · +0 stars on 2026-09-27 (UTC), ≈ 0 by evening
- Star trust: star growth looks organic
- Category: Language models · Language: Python · License: MIT · Created: 2026-09-10 · Last push: 2026-09-10
- GitHub: https://github.com/gauravapiscean/agentic-kv-cache · Page: https://gitnova.dev/en/r/gauravapiscean/agentic-kv-cache

## Useful for

- Test your own KV-cache eviction policy on real agent-session traces
- Reproduce Mooncake's published hit-rate-vs-capacity curve
- Measure how much recompute comes from capacity pressure rather than TTL

## Why it’s here

- The repository is 17 days old and already has 23 stars.
- Hacker News: “LRU is harder to beat than the KV-cache papers suggest” — 108 points, 16 days ago.

## Star trust

Star growth looks organic.

Star-trust labels are heuristics based on the repository’s behavior, not a check of every stargazer.

## Numbers

- Forks: 2
- Issues and pull requests: 0
- Watchers: 0
- Average over the last week: 0 per day
- Usual pace: 1 per day

## Stars per day, last 30 days (oldest → newest, today is partial)

2026-09-30 … 2026-10-29: None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None, None

## Magnitude by day, last 14 days

09-14 1.5, 09-15 1.0, 09-16 1.0, 09-17 0.5, 09-18 0.4, 09-19 0.3, 09-20 0.3, 09-21 0.3, 09-22 0.3, 09-23 0.2, 09-24 0.1, 09-25 0.1, 09-26 0.1, 09-27 0.1

## Hacker News

- LRU is harder to beat than the KV-cache papers suggest — 108 points, 58 comments: https://news.ycombinator.com/item?id=49643543

---

Magnitude (0–10) measures how fast and how unusually interest in a repository is growing right now. It is not a quality score.
Days are UTC. “So far today” is a fact; “expected by the end of the day” is a forecast.
Summaries and use cases are written by an LLM (DeepSeek V4.1 Flash) from the README and may be inaccurate: verify specific claims (benchmarks, speed, hardware) in the repository itself.
Data as of 2026-09-27 13:34 UTC, updated every 30 minutes.