 Alexander Korolev
|
d8dc94acad
release: sm121 0.2.2 — dashboard bench window (params, node, live log, cancel, reopen)
|
1 gün önce |
|
Alexander Korolev
|
fc460279dc
web: cache + settings windows; cache update is one code path
|
4 gün önce |
|
Alexander Korolev
|
396e02e61a
cache update + hf search/card
|
4 gün önce |
|
Alexander Korolev
|
26b02171fc
cache: 'sm121 cache status|doctor' — the shared HF cache, read-only first
|
4 gün önce |
|
Alexander Korolev
|
43d40a41b5
flare driver: orbit node-name mapping, readiness wait, logs attach
|
5 gün önce |
|
Alexander Korolev
|
a9cffba5fa
web: build-version badge under the dashboard title
|
6 gün önce |
|
Alexander Korolev
|
fd30bf160a
web/report: settings button — per-agent client config for a recipe's served model
|
1 hafta önce |
|
Alexander Korolev
|
8a683153dd
bench: add --node <name> — pin the bench target within a cluster (default: rank 0). A serve launched with 'serve --node' lives on a non-rank-0 node (solo Bolt on dgx3 inside 4nodes-switch), and benching the rank-0 head measured a different engine (the live kindling on dgx1 — model_not_available). pick_head tolerates the .lan suffix; web API passes node: None.
|
1 hafta önce |
|
Alexander Korolev
|
8f78bcdc45
serve --cold: recipe-declared cold_args overlay for first-launch safety
|
1 hafta önce |
|
Alexander Korolev
|
0e826d4b35
bench: add --no-think and --timeout passthrough to tool-eval-bench — unblocks valid quality benches for thinking-by-default models (Ling, GLM-4.5-style) whose default reasoning blows the 120s/turn timeout and excludes most scenarios (the 10-16 fp8 Ling run scored 67/100 at 3.1% completion for exactly this reason)
|
1 hafta önce |
|
Alexander Korolev
|
fe1d70ada3
engine: DGPP driver + native deploy (Phase 3) — head-coordinated, generated cluster JSON/.env
|
1 hafta önce |
|
Alexander Korolev
|
5be604ccfe
engine: EngineDriver trait — first-class engine abstraction (Phase 1)
|
1 hafta önce |
|
Alexander Korolev
|
425da04ac3
bench: --label optional (default 'perf' for perf-only), depth suite now opt-in (--depth), parse perf-box caveat asterisks, quiet json warning for perf-only
|
2 hafta önce |
|
Alexander Korolev
|
bb95694de7
report: lazy .log reads (capped head/tail — the multi-MB transcript middle is never parsed; md/json cover the table), text parsers moved to parse.rs; dhat profiler inits in main (dumps on graceful exit); web graceful shutdown capped at 10 s (SSE must not block container stop)
|
2 hafta önce |
|
Alexander Korolev
|
5f1783f3e1
memory: mimalloc allocator feature (add-on builds), lazy report refresh (file-event-driven, fleet cross-check rides the dashboard bus), SSH key cache, graceful web shutdown (SIGTERM) + dhat profiler feature — the Pi OOM-kills (2.7 GiB RSS) traced to musl's never-trimming allocator ratcheting on per-tick transient spikes
|
2 hafta önce |
|
Alexander Korolev
|
f4aee53e0c
single working set: recipes/ + docker-images/ move to the shared repo checkout (../recipes) — dev config points at it via recipe_sources/docker_image_sources; validate gates over every configured source; web hosts the recipe-repo resync loop (--repo-sync-interval, default 10, 0 disables); playwright UI suites run serially (browser lock + split test legs)
|
2 hafta önce |
|
Alexander Korolev
|
bc198f71e1
web: background recipe-repo re-sync loop (--repo-sync-interval N, default 10, 0 disables) — the dashboard stays fresh without a restart; the add-on entrypoint loop becomes redundant
|
2 hafta önce |
|
Alexander Korolev
|
f93a151d40
bench: --no-depth flag — skip the 16K..256K depth suite (quick-grasp campaigns: scenarios + typical-depth perf only)
|
2 hafta önce |
|
Alexander Korolev
|
89ddf56e62
web: secure dashboard auth — Bearer header + session cookie, no token in URLs; --insecure mode
|
2 hafta önce |
|
Alexander Korolev
|
65c1cfcdc8
HA add-on + recipe repos (specs/tech-architecture/ha-addon-recipe-repos.md)
|
2 hafta önce |
|
Alexander Korolev
|
84074f6272
recipe contract gate: sm121 validate + make-check test; archive 22 superseded lanes
|
2 hafta önce |
|
Alexander Korolev
|
3acc8fa855
power on: ensure-powered — nodes believed down with an already-on plug get a 30 s plug power cycle (2-5 s cuts left the PSU rail energized and the machine in soft-off); per-node outcome report becomes the job message; CLI uses a 1.5 s TCP reachability probe, web uses the last snapshot
|
3 hafta önce |
|
Alexander Korolev
|
b70649a547
feat(bench): --ssh-host — run tool-eval-bench on a remote host over SSH
|
3 hafta önce |
|
Alexander Korolev
|
6b16590b1f
sync: one docker save + parallel fabric fan-out (3x on a 4-node set), inner hop key = node-side ~/.ssh/<basename> (mac path broke dgx5-8), known_hosts lives until task end (10G fallback resurrected); sm121 sync-image command
|
1 ay önce |
|
Alexander Korolev
|
0ee032e4e6
bench: depth-suite phase (16K/64K/128K/256K @ c2, 2 repeats) appended to the campaign log; report/tests.rs split (perf_box_tests) for the line cap
|
1 ay önce |
|
Alexander Korolev
|
f39449c4ca
bench: --parallel passthrough + config C bench 166/184 (90/100, two-stream)
|
1 ay önce |
|
Alexander Korolev
|
bc269457c2
bench(exl3-staged): 92/100 declarative-staging verification + bench --tokenizer passthrough
|
1 ay önce |
|
Alexander Korolev
|
58c6892537
staging P1: declarative [[staging]] hub-mirror, ledger-verified
|
1 ay önce |
|
Alexander Korolev
|
45462379ae
bench: re-attach the perf table the --json-file path drops; fix aiden recipe toml
|
1 ay önce |
|
Alexander Korolev
|
51b7404aa8
feat: serve --hf-cache <path> overrides the host-side HF cache dir per serve (config hf_cache_dir stays the global knob; mutual exclusion with --mirror)
|
1 ay önce |
