Three pages described the fleet; two of them still described the pre-September arrangement.
Re-read live on 2026-10-21 (`/v1/models`, `systemctl`, `lspci`, `nproc`, `resolvectl`):
- `system/network.md` called Dankclaw the "large inference model" box and Dankserver the
"small" one — that is the topology from before the RTX 3090 moved in September. The served
ids today are `frank-archon` on Dankserver:7770 (the main model) and `frank-dragoon`
(:7770) + `frank-zealot` (:7771) on Dankclaw. The gateway and the DNS pin were re-read too.
- `system/dankserver.md` claimed `frank-medium` is *masked*: the unit file does not exist
(`is-enabled` answers `not-found`), which is a different fact with a different remedy. And it
claimed 8C/16T: this 7700X exposes **8 threads** (`nproc` = 8), so anything sized for 16 is
sized for a machine that is not this one.
- `system/dankclaw.md` transcribed the engine's slot/KV/context math — `2×262,144 slots over
one 524,288 pool, unquantized f16 K+V, MTP off` — while the live unit says `3x200704 slots
over one 602112 pool, f16 K + q8_0 V, MTP armed`, and dragoon's ctx had moved too. That math
belongs to the serverAppConfigs runbook the page already names: the unit's `ExecStart` is a
wrapper script, so any copy here rots on the next engine change. The page now keeps the
contract — id, unit, port, engine family, which GPU, and how to prove it — and disowns the
numbers, with the reason stated so the next agent does not re-transcribe them.
Also recorded, because it is how this audit nearly missed a stopped service: `curl :7770/v1/models`
answered normally while `frank-dragoon.service` sat `inactive` with `Result=exit-code` — the
container is running outside systemd. A serving port is not a healthy unit. (Flagged for Frank
separately; no production unit was touched.)
Refs #64
type, title, description, tags, generated, verified, status
type
title
description
tags
generated
verified
status
reference
Dankserver
Radiance inference server at 192.168.0.235:7770, serving frank-archon (Qwen3.8-Flash-Next iq4r, 2x R9700 TP=2, 200,000-token context).
network
llamacpp
inference
server
by
chief-designer/agent
2026-10-21
stable
Dankserver
Primary inference server for the Frank Program. Serves frank-archon, the
main agent model, from a pair of Radeon AI PRO R9700s (tensor-parallel 2).
The RTX 3090 that used to live here was removed 2026-09-15 and now runs on
Dankclaw.
200,000 tokens (harness models.yml; served pool may differ — see runbook)
frank-medium (the former dual-GPU mid tier) no longer exists on this box —
systemctl list-unit-files | grep frank- returns only frank-archon.service. It was retired,
not masked: on 2026-10-21 systemctl is-enabled frank-medium answers not-found.
Usage
Dankserver carries the main-session model. Subagent role overrides in
agent/models.yml route here for task and reviewer; scout/sonic/designer
route to Dankclaw.
# Health check
curl http://192.168.0.235:7770/health
# Quick test
curl http://192.168.0.235:7770/v1/chat/completions \
-H "Content-Type: application/json"\
-d '{"model":"frank-archon","messages":[{"role":"user","content":"Hello"}],"max_tokens":32}'
Hardware (verified live 2026-10-21)
Property
Value
CPU
AMD Ryzen 7 7700X — 8 cores, 8 threads exposed (nproc = 8; SMT is not enabled, so do not plan for 16)
GPU
2× AMD Radeon AI PRO R9700 (Navi 48, 32 GiB each)
RAM
246 GiB
Swap
effectively none (~0.1 GiB) — the 64 GiB swapfile is on Dankclaw, not here
Storage
7× 18.2 TB HDD + 1.8 TB NVMe (layout owned by the deploy repo)
OS
Ubuntu 26.04.1 LTS, kernel 7.0.0-38-generic
The old page (verified 2026-08-28) listed an RTX 3090, 60 GiB RAM, and a
64 GiB swapfile here; all three were superseded by the 2026-09 hardware swap.
Related Knowledge Bases (sibling project)
serverAppConfigs — deployment repo + vault, on this workstation at
~/Projects/serverAppConfig/ (docs vault: ~/Projects/serverAppConfig/serverAppConfigs-docs/).
Podman/quadlet deploy config for both servers plus the LLM inference fleet.
The live frank-archon runbook (engine flags, KV pool sizing, restart
procedures) is LLM/frank-archon-radiance-qwen38-iq4r.md in that vault —
it, not this page, is the deploy-side source of truth.
See Also
dankclaw — the other inference box (dragoon + zealot)
Frank
54bb0202c2