Skip to content

fix(serve): K1 — GPU KV = model context within VRAM, not a fixed 4096 (#4603) - #4630

Closed
noahgift wants to merge 52 commits into
mainfrom
devont/k1-serve-kv
Closed

noahgift wants to merge 52 commits into
mainfrom
devont/k1-serve-kv

Commits

  1. Commits on Sep 28, 2026

  2. Commits on Sep 29, 2026

  3. Commits on Oct 1, 2026

  4. Commits on Oct 2, 2026

  5. Commits on Oct 3, 2026

  6. Commits on Oct 4, 2026