dank/serverAppConfigs
1
0
Fork 0
Code Issues 6 Pull Requests Actions Packages Projects Releases Wiki Activity
Files
c2116658f89fcb1201dd42dcfdc1d1f68e65d310
serverAppConfigs/shared/homepage/services.json
T
Frank c2116658f8 feat(frank-medium): cut the endpoint over to radiance TP=2 
The conversation battery (scripts/bench-conversation.py, salted, same 10
prompts, 2026-09-20) measured three arms on the two R9700s:

  arm I  llama.cpp layer split 32:33 + DFlash2, Q8_0  geo decode  45.7 t/s, TTFT 0.72 s, burst-4  71.7 t/s
  arm L  llama.cpp tensor split 1:1 (RCCL), no drafter geo decode 30.5 t/s, TTFT 0.71 s, burst-4  54.3 t/s
  arm R  radiance vLLM TP=2 + DFlash2-FP8, MXFP4      geo decode  95.8 t/s, TTFT 0.21 s, burst-4 259.4 t/s

Radiance wins every axis, so it takes the alias. The board swap is what made
arm R possible: TP=2 was archived on 2026-09-16 because the second card sat
behind B650 chipset bridges with AtomicOpsCap: Routing-, and the ProArt
B850-Creator put both cards on CPU GPP root ports (Gen5 x8, Routing+, ReqEn+).
The R4D one-shot all-reduce needs no RCCL collective anyway.

  - un-retire dankserver/frank-medium/: the TP=2 quadlet (banner rewritten —
    it is the deployed shape now), radiance-entrypoint.sh, the healthcheck.
    The replica pair and its affinity proxy stay archived; one process owns
    both cards now.
  - retire scripts/frank-medium.{service,env} + serve-frank-medium.sh and
    bench-spec-ab.py to scripts/.AGENTTRASH/20260920-llama-layersplit/ as the
    rollback path. build-llama-rdna4.sh and download-frank-medium-gguf.sh
    stay: they are the tooling that rollback needs.
  - deploy-remote.sh: deploy_llm_medium now gates on the image (pull once,
    re-tag to the localhost name the quadlet resolves), the MXFP4 checkpoint,
    the drafter, the pinned patch clone and the patched libr4d; retires the
    llama root unit into the home trash instead of relinking it; then
    install_quadlet + wait_container_healthy 1800s (first start compiles).
    retire_vllm_medium_quadlets becomes retire_vllm_replica_quadlets — the
    a/b/proxy trio only, since frank-medium is installed again.
  - planner + test-changed map dankserver/frank-medium/** to frank-medium
    and its seam; the scripts-trio mapping is gone with the files.
  - tests/frank_medium_unit.bats rewritten for the container contract: boot
    persistence of a GENERATED unit (no enable on this host's v1 generator),
    TP=2 with the atomics-era NCCL workarounds kept out, the DFlash2 depth vs
    the MXFP4 decode-kernel row band, the KV pin's coupling to the shape it
    was calibrated at plus pool/VRAM arithmetic, the endpoint/tool/reasoning
    contract, and the archive invariants.
  - homepage card + docs updated, and they were stale twice over: they still
    described the ajgazin Swift 35:30 / 393k payload, not the deployed shape.

Known regression, documented in the unit: vision. The MXFP4 checkpoint is
language_model_only and the launcher loads no projector, so images are a 400
here; the harness's vision role already routes to frank-small.
2026-09-20 18:15:27 -07:00

366 lines
8.9 KiB
JSON
Raw Blame History

[
{
"name": "SearXNG",
"category": "user",
"description": "AI-free, VPN-routed search engine, fully hosted on Frank's server.",
"url": "http://search.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:8081",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "OpenWebUI",
"category": "user",
"description": "AI Chat/Search interface",
"url": "http://ai.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:3005",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Gitea",
"category": "user",
"description": "Git code hosting and review platform.",
"url": "http://gitea.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:3000",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Calibre-Web",
"category": "user",
"description": "E-book library.",
"url": "http://books.franksdank.cloud",
"docs": "",
"probe": "192.168.0.235:8083",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "RomM",
"category": "user",
"description": "ROM and emulator library.",
"url": "http://romm.franksdank.cloud",
"docs": "",
"probe": "192.168.0.235:8085",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Copyparty",
"category": "user",
"description": "Direct downloads from Frank's server.",
"url": "http://files.franksdank.cloud",
"docs": "",
"probe": "192.168.0.235:3923",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Plex",
"category": "user",
"description": "Closed source media server with apps for everything.",
"url": "http://plex.franksdank.cloud/web/index.html",
"docs": "",
"probe": "192.168.0.235:32400",
"method": "tcp",
"wip": false,
"showLink": true
},
{
"name": "Jellyfin",
"category": "user",
"description": "Open source media server.",
"url": "http://jelly.franksdank.cloud",
"docs": "",
"probe": "192.168.0.235:8096",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Seerr",
"category": "user",
"description": "Download Requests",
"url": "http://seerr.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:5055",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Marreta",
"category": "user",
"description": "Clean, ad-free article reader. Paste any URL, get just the article.",
"url": "http://marreta.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:8082",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Sonarr",
"category": "admin",
"description": "TV series library and automation.",
"url": "http://192.168.0.112:8989",
"docs": "",
"probe": "192.168.0.112:8989",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Hermes",
"category": "admin",
"description": "The agent gateway.",
"url": "",
"docs": "",
"probe": "127.0.0.1:8787",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "Grafana",
"category": "admin",
"description": "Dashboards.",
"url": "http://grafana.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:3002",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Prometheus",
"category": "admin",
"description": "Metrics collection.",
"url": "http://prometheus.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:9090",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Loki",
"category": "admin",
"description": "Log storage.",
"url": "http://loki.franksdank.cloud",
"docs": "",
"probe": "127.0.0.1:3100",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "node-exporter",
"category": "admin",
"description": "Host metrics.",
"url": "",
"docs": "",
"probe": "127.0.0.1:9100",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "alloy",
"category": "admin",
"description": "Log shipping.",
"url": "",
"docs": "",
"probe": "127.0.0.1:12345",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "alertmanager",
"category": "admin",
"description": "Alert delivery to Signal.",
"url": "",
"docs": "",
"probe": "127.0.0.1:9093",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "Caddy",
"category": "admin",
"description": "The reverse proxy.",
"url": "",
"docs": "",
"probe": "127.0.0.1:80",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "dnsmasq",
"category": "admin",
"description": "LAN name resolution.",
"url": "",
"docs": "",
"probe": "127.0.0.1:53",
"method": "tcp",
"wip": false,
"showLink": false
},
{
"name": "Gluetun",
"category": "admin",
"description": "The outbound VPN tunnel.",
"url": "",
"docs": "",
"probe": "127.0.0.1:51820",
"method": "tcp",
"wip": true,
"showLink": false
},
{
"name": "frank-large",
"category": "user",
"description": "The big brain. Qwen3.8-Flash-Next w4b on halogen-flash-server, running on Dankclaw's Strix Halo iGPU. Text-only; holds two 200k-token conversations resident.",
"url": "",
"docs": "/docs#models",
"probe": "127.0.0.1:7770",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "frank-medium",
"category": "user",
"description": "The 27B generalist. Qwen3.8-27B-Uncensored at MXFP4 on vllm-radiance, one tensor-parallel process across Dankserver's two R9700s with a DFlash2 drafter; 262k-token context shared by up to eight conversations.",
"url": "",
"docs": "/docs#models",
"probe": "192.168.0.235:7770",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "frank-small",
"category": "user",
"description": "The vision path. HauhauCS Gemma4-12B QAT (Balanced, Q4_K_M) with MTP + ngram spec-decode on Dankclaw's RTX 3090. Sees images; 1M-token pool across four conversations.",
"url": "",
"docs": "/docs#models",
"probe": "127.0.0.1:7771",
"method": "http",
"wip": false,
"showLink": false
},
{
"name": "mariadb",
"category": "admin",
"description": "RomM's database (Dankserver). Not probeable from the monitor: 3306 is container-internal, and publishing a database port on the LAN is not a trade this fleet makes. RomM's own card covers the stack.",
"url": "",
"docs": "",
"probe": "192.168.0.235:3306",
"method": "tcp",
"wip": true,
"showLink": false
},
{
"name": "valkey",
"category": "admin",
"description": "SearXNG's cache/limiter (Dankclaw). Not probeable: 6379 is container-internal. SearXNG's own card covers the stack.",
"url": "",
"docs": "",
"probe": "127.0.0.1:6379",
"method": "tcp",
"wip": true,
"showLink": false
},
{
"name": "Prowlarr",
"category": "admin",
"description": "Indexer manager.",
"url": "http://prowlarr.franksdank.cloud",
"docs": "",
"probe": "192.168.0.112:9696",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Radarr",
"category": "admin",
"description": "Movie library and automation.",
"url": "http://radarr.franksdank.cloud",
"docs": "",
"probe": "192.168.0.112:7878",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Readarr",
"category": "admin",
"description": "Book library and automation.",
"url": "http://readarr.franksdank.cloud",
"docs": "",
"probe": "192.168.0.112:8787",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Lidarr",
"category": "admin",
"description": "Music library and automation.",
"url": "http://lidarr.franksdank.cloud",
"docs": "",
"probe": "192.168.0.112:8686",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "qBittorrent",
"category": "admin",
"description": "Torrent client.",
"url": "http://qbittorrent.franksdank.cloud",
"docs": "",
"probe": "192.168.0.112:8080",
"method": "http",
"wip": false,
"showLink": true
},
{
"name": "Postgres",
"category": "admin",
"description": "Authentik's and Gitea's databases (Dankclaw). Not probeable: 5432 is container-internal. Their apps' cards cover the stacks.",
"url": "",
"docs": "",
"probe": "127.0.0.1:5432",
"method": "tcp",
"wip": true,
"showLink": false
},
{
"name": "ComfyUI",
"category": "admin",
"description": "Node-based image generation.",
"url": "",
"docs": "",
"probe": "127.0.0.1:8188",
"method": "http",
"wip": false,
"showLink": false
}
]
Reference in New Issue View Git Blame Copy Permalink