Pull requests / #2834
#2834 docs: --prompt-cache-every is not a piece-size dial here (the fork's is)
open · @gopinath87607 · 0 Kommentare · Auf GitHub
Beschreibung
## What this changes A docs bullet for `--prompt-cache-every`, in `docs/DETAILS.md`: the knob is **not** a piece-size dial in this engine. In the Maya checkout it is also a wall-time dial because there its marks <em>bound</em> the segments the prompt path reads, and every segment pays a fill and a drain of the part chain. What the bullet records: - The Maya checkout's numbers for its `STRATA_GLM_CKPT_EVERY` (same 16384 default), local commit `3095719` (2026-10-15), on a 100,535-token fresh read (Maya-L split, 8192-token chunks, one 580-token warm prompt before each arm, engine restarted between arms): `16384` cut the read into 5 pieces read in 148.8 s, `32768` into 4 in 134.9 s (-9.3%), `0` into one of 13 chunks in 93.2 s (-37.4%) — that checkout's numbers on its rig, a piece fitting ~14.6 s + 6.5 s a chunk there. - Why they do not carry over here: a periodic mark is saved at a chunk boundary and the run carries on, so no piece ends at it. What cuts this engine's read into runs is the cut list — a re-read (`STRATA_CKPT_REREAD`), the system-prompt root, a message boundary (`STRATA_CACHE_MESSAGE_BOUNDARY`), the start of the last turn and a pinned prefix — and each of those is one fill and drain. - So `--prompt-cache-every 0` here skips the periodic marks and the near-tail one (a `--prompt-cache-tail` mark needs `--prompt-cache-every` above 0), while the root, the last turn, a message boundary and a pin are still saved — the same trade the Maya checkout documents for its `0`: a client that rewrote a turn *before* the last one mounts a farther mark. The lever for a fresh long read here is those cuts, not this cadence. ## What it leaves alone Documentation only: not one line of code or default changes. Every name and flag the bullet cites was checked against the tree it documents (`--prompt-cache-every`, `--prompt-cache-tail`, `STRATA_CKPT_REREAD`, `STRATA_CACHE_MESSAGE_BOUNDARY`). ## Measurements The numbers above are quoted with what they were measured on (the Maya checkout's rig, Maya-L split, the warm prompt, the restarts between arms). No number is claimed for this engine. ## Checks - Docs-only change; no build to run. The rendered Markdown is the only output. 🤖 Generated with [Claude Code](https://claude.com/claude-code)
Mehr auf der Site
Links zu Install, Modellen, Releases.