Skip to content

AtomicBot-ai/atomic-llama-cpp-turboquant

View on GitHub ↗Related repositories →

llama.cpp fork with TurboQuant WHT-rotated KV cache & weight compression + TQ3_RVQ: 3.44 bpw residual vector quantization for LLM weights — WHT + iterative RVQ TQ3_1S at 10.8% smaller model size.

active 2026-03-302026-08-07 (UTC)

Complete coverage26,532 / 26,532 hourly files (100%) · 2 absent upstream2023-08-152026-08-24 (UTC)
Events
83
Pushes
27
Pull requests
20
Issues
0
Stars
24
Forks
1

Activity over time

Daily event counts in the loaded window

Line chart, 131 days from 2026-03-30 to 2026-08-07. Pushes: 27 total, peak 5 in a day. Pull requests: 20 total, peak 16 in a day. Issues: 0 total, peak 0 in a day. Comments: 2 total, peak 1 in a day. Stars: 24 total, peak 4 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
github-actions[bot]180180
Ooooze121200
Vect0rM9900
worthant6510
dtorey-d1100
onyoureyes1001
WillowOneVision1001
sujitvasanth1010

Recent activity

Latest issues, pull requests and releases

  • Pull request#33worthant2026-07-17 13:11
  • Issue comment#19WillowOneVision2026-05-21 23:27
    Phase C.2 dispatch behavior: MTP+mmproj coexistence behind --allow-mtp-with-mmproj (5th first-in-world)
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Pull request#15github-actions[bot]2026-05-17 01:33
  • Releasegithub-actions[bot]2026-05-12 19:13
    TurboQuant macOS ARM64 (b1a7d71)
  • Issue comment#5onyoureyes2026-05-12 10:51
    Repro: MTP path on CUDA aborts at fattn.cu:109 (DKQ=512) for Gemma 4 — Blackwell sm_120 + Ampere sm_86
  • Pull request#8github-actions[bot]2026-05-11 02:15
  • Pull request#8github-actions[bot]2026-05-11 02:15
  • Pull request#7sujitvasanth2026-05-11 00:57

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 24 stars here means stars gained during the window, not the repo's star count.