d710e56aca
Split the 53 over-threshold dated log entries (Recent decisions, Tried and abandoned) into per-entry persistent-memory.d/<slug>.md detail files, leaving one-line pointers in the index; the 10 short entries stay inline. Startup index drops 60,527 -> 25,256 bytes (492 -> 270 lines); entry bodies move verbatim to on-demand detail files, so a fresh session loads ~25 KB instead of ~60 KB and pulls a detail file only when its pointer is relevant. Top matter (Repo purpose, Tools & conventions, Current state) is unchanged; both archival back-references preserved. CLAUDE.md persistent-memory section now documents the index<->detail read discipline (read the index, pull details on demand, never bulk-read the dir, commit both together). Auto-archival still held every dated entry back (all <30 days old); the July burst begins aging past the 30-day guard ~2026-07-31.
1.3 KiB
1.3 KiB
[2026-07-14]NVFP4+MTP fast char-rp-reasoning seat LANDED + LIVE + gateway-repointed + VRAM-tuned. Modelopt-format re-quant made MTP work. The load-crash root cause = vLLM 0.24 does NOT propagate modeloptexclude_modulesto the spec-decode DRAFT model → the bf16 mtp head gets quantized → shape crash; NO checkpoint config fixes it (is_layer_skippedis exact-membership, and the drafter never sees the mtp excludes) → workaround = a mountedsitecustomizethat force-skipsmtp.*inis_layer_skipped(upstream vLLM bug to file). Productionized as compose stackheretic2-charrp-reasoning(:8018, workaround baked in). Gatewaychar-rp-reasoningalias fixed: repointed off the stale GGUF served-namedeckard-pkd-27b, addedenable_thinking:true, droppedmin_p(MTP-incompatible), canonical samplers temp1.0/top_p0.95/top_k20. Rebalanced GPU0 (gen 0.37→0.30/16-seq/256K + reasoning 0.39/16-seq/192K+MTP + char-rp 128K, 2.7GB free). All 4 gateway roles verified; vLLM reasoning-parser confirmed leak-free (unlike the GGUF budget-forcing). Full record + the 4 quant landmines indocs/runbooks/heretic2-nvfp4-mtp-seat.md; committed982c319. Open (non-blocking): brokkr P00 (seat is live ahead of it), retire the stopped GGUF reasoning seat, file the vLLM bug.