Skip to content

microsoft/RetrievalAttention

View on GitHub ↗Related repositories →

Scalable long-context LLM decoding that leverages sparsity—by treating the KV cache as a vector storage system.

active 2025-05-152026-03-11 (UTC)

Complete coverage26,530 / 26,530 hourly files (100%) · 2 absent upstream2023-08-152026-08-24 (UTC)
Events
131
Pushes
8
Pull requests
11
Issues
3
Stars
75
Forks
13

Activity over time

Daily event counts in the loaded window

Line chart, 301 days from 2025-05-15 to 2026-03-11. Pushes: 8 total, peak 1 in a day. Pull requests: 11 total, peak 3 in a day. Issues: 3 total, peak 1 in a day. Comments: 15 total, peak 4 in a day. Stars: 75 total, peak 7 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

ContributorContributionsPushesPRsComments
dependabot[bot]12057
Komorebi66012434
baotonglu5400
iofu7283030
framolfese3003
liurt12181000
microsoft-github-policy-service[bot]1001

Recent activity

Latest issues, pull requests and releases

  • Pull request#17Komorebi6602026-02-22 07:36
  • Issue#16baotonglu2026-02-22 07:35
    Where can I find the code for the clustering algorithm?
  • Issue comment#16Komorebi6602026-01-23 03:16
    Where can I find the code for the clustering algorithm?
  • Issue comment#14Komorebi6602025-09-19 00:39
    Prefilling Latency
  • Issue comment#15framolfese2025-09-17 13:48
    RuntimeError: Event device does not match recording stream's device
  • Issue comment#15Komorebi6602025-09-17 10:23
    RuntimeError: Event device does not match recording stream's device
  • Issue comment#15framolfese2025-09-16 16:49
    RuntimeError: Event device does not match recording stream's device
  • Issue comment#13framolfese2025-09-02 08:48
    [Environment Setup] Failed to install dependencies
  • Issue comment#6dependabot[bot]2025-08-18 10:01
    Bump vllm from 0.6.2 to 0.9.0
  • Issue comment#7dependabot[bot]2025-08-18 10:01
    Bump torch from 2.4.0 to 2.7.1
  • Pull request#11Komorebi6602025-08-18 10:01
  • Issue comment#12dependabot[bot]2025-08-18 10:01
    Bump transformers from 4.49.0 to 4.53.0
  • Pull request#12Komorebi6602025-08-18 10:01
  • Issue#10Komorebi6602025-08-18 10:00
    [feat] New Cache Evict Policy
  • Issue comment#13Komorebi6602025-08-18 09:53
    [Environment Setup] Failed to install dependencies
  • Pull request#12dependabot[bot]2025-08-06 20:05
  • Issue comment#11microsoft-github-policy-service[bot]2025-07-25 13:18
    fix issue #10
  • Issue#10liurt12182025-07-25 12:51
    [feat] New Cache Evict Policy
  • Issue comment#8dependabot[bot]2025-07-08 18:06
    Bump transformers from 4.49.0 to 4.51.0
  • Pull request#9dependabot[bot]2025-07-08 18:06
  • Pull request#8dependabot[bot]2025-07-08 18:06
  • Pull request#8dependabot[bot]2025-07-08 05:48
  • Pull request#6dependabot[bot]2025-05-28 18:04
  • Issue comment#3dependabot[bot]2025-05-16 16:28
    Bump transformers from 4.49.0 to 4.50.0
  • Pull request#3iofu7282025-05-16 16:28

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 75 stars here means stars gained during the window, not the repo's star count.