Skip to content

為了加快 LLM 的推理速度並增強 LLM 對關鍵資訊的感知,可以壓縮提示和 KV-Cache,從而實現高達 20 倍的壓縮,同時將性能損失降至最低。To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

active 2023-10-09 → 2026-09-18 (UTC)

Complete coverage27,286 / 27,289 hourly files (100%) · 2 absent upstream2023-08-15 → 2026-09-25 (UTC)
Events
6.6K
Pushes
103
Pull requests
116
Issues
199
Stars
5.3K
Forks
334

Activity over time

Daily event counts in the loaded window

Line chart, 1076 days from 2023-10-09 to 2026-09-18. Pushes: 103 total, peak 8 in a day. Pull requests: 116 total, peak 6 in a day. Issues: 199 total, peak 5 in a day. Comments: 355 total, peak 14 in a day. Stars: 5,316 total, peak 410 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

Recent activity

Latest issues, pull requests and releases

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 5,316 stars here means stars gained during the window, not the repo's star count.