CTX 262144 / PARALLEL 2 = two 131,072 conversations; SPEC_TYPE=none strips both speculative flags in the serve script (ngram-mod skipped on purpose — on a qwen35 hybrid it disables recurrent-state rollback); ubatch 1024->512 claws back ~0.3 GiB. Budget: non-KV ~15.5 GiB + two q8_0 slots 8,704 MiB lands AT the card — the 2-3 GiB floor is knowingly violated by explicit order; nvidia-smi free at boot is the measured truth. Seam retargeted: shape (2 slots, 131072/slot), budget (two fit, three don't), spec-off (absence of both flags), unit Description.
Run 3 rewrote the staged index THROUGH the relative symlink into the raw
checkpoint: the raw index.json was restamped with the stacked layout
(605 KB smaller), the raw verify then failed 143/144 and the overnight
run died before stacking. Two fixes:
- stack-qwen38-official-fp8.py: index/config writes are now temp+os.replace
(an atomic swap replaces the symlink itself; the raw tree stays pristine).
- build-frank-primary-rad.sh: the stage's index.json is COPIED like
config.json; only shards stay relative symlinks.
- release the numpy frombuffer view before closing the source mmap
(Python 3.14 BufferError: cannot close exported pointers exist).
- stacked scale planes rename: checkpoint.cpp scale_name() appends
_scale_inv to the STACKED tensor name verbatim (no .weight segment) -
...experts.gate_up_proj_scale_inv, not ...gate_up_proj.weight_scale_inv.
Boot ledger lines, 19,404 used / 4,721 MiB free (the q8_0 proof), decode arms (66.1 t/s @0.8, 66.9 @0, 58.9 @recipe 1.0 with acceptance 52.1%/52.2%/44.8%), vision Red probe, reasoning_effort xhigh/high contract, e2e 12-crawl transient + full-gate green rerun. Unit Description updated in-repo; the relinked unit picks it up at the next deploy/restart (no bounce for cosmetics). 360/360 suite green.
The parallel session's uncommitted swap, committed at Frank's direction
(commit whatever's in the working tree):
- llama.cpp master on the RTX 3090 now serves ukisai's
Swift-1.5-Qwen3.8-27B GSQ-RCO IQ3_S WITH the in-file MTP head, ONE
131,072-token slot at Frank's 128k cap, q8_0 K + q8_0 V, vision through
the base-model projector; MiMo-V2.6 Q6_K (2x200704 f16 KV) retired to
rollback target.
- env/serve/download + seam twins updated together (deployer payload gates
rode the previous commit's deploy-remote.sh).
- docs: new swift15 runbook; MiMo Q6K runbook retired; AGENTS.md topology
and both vault indexes follow (root index ACTIVE/RETIRED flip, test 135
seam green).
- comfyui: two qwen-image t2i workflow drops under dankclaw/comfyui/
workflows/ (12 KiB).
Replace the FreeToken PR587 endpoint (single-GPU: its TP=2 dies in the
pynccl.cu ROCm JIT) with the radiance engine on both R9700s, TP=2 via
libr4d's fused one-shot all-reduce. Official Qwen/Qwen3.8-Flash-Next-FP8
payload at native block-FP8 precision:
- scripts/build-frank-primary-rad.sh: staging (quantization_config drop =
plugin SELECTION law) -> checkpoint reshape -> rad-convert with ONE
--quant rule (PLE rtn fixed-scale); plan-only first.
- scripts/stack-qwen38-official-fp8.py: the two value-preserving reshapes
(experts byte-stacked into the plugin's stacked layout; PLE's 128
scaleless fp8 shards decoded to one bf16 shard with the checkpoint's own
scale; the rtn rule re-encodes at that same scale - codes round-trip
exactly and land the one-element scale plane the plugin declares).
- scripts/hfget.sh + verified download (131/131 shards sha256-match).
- froggeric v22.5 chat template vendored (--override-chat-template).
- serve script (pinned ledger: 2x40 GiB pools + 47.7 GiB PLE + prefix host),
TTM preflight fixed for kernels with pages_limit only (no pages_max -
an absent knob is skipped, not read as zero); pages_limit raised to
52,428,800 pages = 200 GiB on the host pre-cutover.
- root unit, relinked; deployer gates image digest + .rad floor 158,000 MiB
+ health; prometheus re-adds the /metrics job.
- rollback: tag frank-primary-freetoken-pr587-20261005 (FreeToken trio +
Dockerfile); payload + image stay on disk until the window closes.
deploy-remote.sh also carries the parallel session's frank-small payload
gates (swift15) - committed here per Frank's direction to commit the tree.
FreeToken's /health returns 200 with {"status":"loading"} while the
engine loads — the FastAPI front end binds before the scheduler is up.
Measured 2026-10-06: the deploy gate went green after 6s while 131 shards
were still paging in (vLLM 503'd during load, which made the bare curl
honest there). The frank-primary gate now loops until status:ok; the
MainPID stray-port guard is unchanged.
Measured first boot: the auto attention backend is qsa_sparse, which
overrides page size to 64 and refuses --num-tokens 300000 ('nearest valid
values: 299968 or 300032'). 299,968 x 24 KiB = 7.0 GiB still clears the
177,356-token harness worst case and one native 262,144 conversation.
Replaces the tcclaviger vLLM fork with FreeToken at PR #587 head 77a7201
built on dankserver into localhost/frank-primary-freetoken:pr587-77a7201
(scripts/docker/frank-primary-freetoken.Dockerfile: digest-pinned
rocm/pytorch rocm7.14 torch-2.11.0 base, libhiprtc dev symlink, curated
--no-deps install per docs/install_amd.md).
Shape (first-boot conservative): ONE R9700 (second card idle;
--tensor-parallel-size exists for future 2-card tuning), KV pool
--num-tokens 300000 (7.2 GiB at 24 KiB/token; clears the 177,356-token
harness worst case and one native 262,144 conversation), MoE LRU
--moe-cache-rate 0.10, --max-running-requests 2, --max-output-tokens
36864. Parsers AUTO: thinking -> reasoning_content natively (models.yml
flips reasoningContentField on the harness side). Experts + PLE pin
~170 GiB host RAM — root unit keeps LimitMEMLOCK=infinity.
Container identity renamed to the unit name (Frank's call): frank-primary;
vLLM-era frank-primary-qwen joins the legacy cleanup list. Deployer image
gate becomes podman image exists (a local build carries no RepoDigests;
the old digest grep refused exactly this shape). Rollback tag:
frank-primary-vllm-shapeG-20261003.
- FROM pinned to sha256:a223aee1... (the pulled rocm/pytorch rocm7.14
pytorch 2.11.0 manifest); a re-pushed tag cannot change the stack.
- The pip-ROCm torch ships libhiprtc.so.7 only; the PR's _cpu_moe links
-lhiprtc and ld needs the unversioned symlink in the already-linked
_rocm_sdk_core/lib (first build died: cannot find -lhiprtc).
The rogue session's swap left this env reprice uncommitted in the tree
(2026-10-05): Ornith is a 33-block qwen35 hybrid with NINE full-attn
layers (18 KiB/token/half), +1,900 MiB of weights vs MiMo Q6_K — f16/f16
lands ~379 MiB free, a dead boot; pool-stepping cannot rescue it because
the compaction invariant needs slot >= 191,148. The only lever that
keeps 2 x 200,704 AND the 2-3 GiB floor is q8_0 V (~2.87 GiB free). K
stays f16. Deployer gate + seam already retargeted in 8f53df7; fetch
script manifest already flipped in 3ee88d7. MEASURE after boot; under
1.5 GiB free, UBATCH_SIZE down to 1024, never the window.
The rogue session's Ornith commits (3ee88d7/77decb5) swapped the env and
fetch script but left the deploy handler gates and the seam asserting
the MiMo-Q6K paths — the seam failures 195-203 and a deploy that gates
a file the unit no longer reads. Flip both to Ornith-Q8 (Q8_0 GGUF +
BF16 mmproj), keeping the anti-drift twin structure intact.
The 11-admin spec walked links that terminate on FrankBook-Pro
(192.168.0.112), which sleeps when idle — every deploy gate timed out
there (measured 2026-10-05, two deploy windows). Split: the frankbook
links move to tests/e2e/specs/19-frankbook.spec.ts under a frankbook
project that run.sh excludes by default and deploy.sh triggers ONLY
when a frankbook blast-radius path changed (the spec, frankbook-pro/,
the homepage registry, the Caddyfile arr vhosts). deploy.sh computes
the trigger from the same BASE..HEAD revision range the plan uses.
The first constants were wrong in three ways, found while resuming the
checkpoint (the file was COMPLETE; the manifest was not):
- checkpoint is 35,865,565,184 B / LFS sha256 274c3fc7... (not
35,880,329,102 / 40874e7a...), 1352 tensors per the card;
- the tokenizer is a FLAT tokenizer/ DIRECTORY of 5 files mounted whole
at /tokenizer (model-card docker-run parity; the engine's
HALOGEN_TOKENIZER default is the directory) — there is no
frank-large.tokenizer.gguf;
- plain git blobs (vocab/merges/jinja/config) carry no LFS oid, so the
fetch script size-checks them and sha-verifies only the two LFS
objects (checkpoint, tokenizer.json).
Gates flipped together (serve + deployer twins): tokenizer dir must be a
real directory and every named file a regular file, no symlinks (HF
snapshot links dangle in the mount). Fetch made idempotent: a file at
the published size skips the transfer (resuming it is a CDN 416 under
set -e). Seam rewritten to the same contract, 17/17.
curl drops the Range header across the HF /resolve/ 302, so '-C -'
against the resolve URL pulls the 1 KB redirect body and exits 0 with
the file still 14 MiB short (measured 2026-10-05). Resolve the
effective URL first; the Xet CDN honors Range.
Runtime swap ROCmFPX llama.cpp -> peonist-ai/halogen (image digest
dc0a39a0, tag 0.1.4), serving qwen3.8-27b-p1w4d-d2.hgn (33.4 GiB, the
.hgn format loads in no other engine) with HALOGEN_DRAFTER=2 (DFlash2)
on ONE slot of 200700 ctx: slots>1 disables the drafter (engine law).
Accepted: TEXT-ONLY (vision encoder ignored; frank-small owns images)
and the censored base (rollback = revert this commit; payload, fork
build and build-rocmfpx.sh pin stay on disk).
- serve script: podman run under the root unit, exec podman wait so
Restart=always is real; gates image digest, checkpoint byte-count,
flat regular-file tokenizer, GPU nodes; engine stdout to log.
- deploy handler: gates the artifacts, sweeps the legacy bare
llama-server + old container name by NAME (never pkill llama-server:
frank-small), retires the 2026-09 quadlet, port cutover 7770.
- fetch script: -C - resume, -f, byte count, sha256, --verify-only.
- seam rewritten (17 tests): halogen is now the deployed runtime.
BATCH_SIZE 2048 -> 4096, UBATCH_SIZE 1024 -> 2048.
Why the ubatch move is right on THIS card: the prefill curve is
card-specific. On the 2x R9700 layer-split sibling it peaks at ubatch 512
and every step up costs GiB for no speed; on this CUDA 3090 it climbs
monotonically to 2048 (710 -> 1188 t/s on the 35B MoE sweep, 1944 -> 2423
t/s on the Ornith 8k sweep) with decode flat to under 1 t/s on every rung.
The logical batch is free either way: two sweeps agree that at a fixed
ubatch, 2048/4096/8192 give identical prefill speed and identical peak
VRAM, because the compute buffers are sized by the ubatch alone.
The risk is all in the ubatch, and it is documented rather than assumed:
the step cost ~1,244 MiB of peak graph VRAM on the 35B MoE, the shape this
replaces is 2,815 MiB free, and the Q5_K_M predecessor recorded "ubatch
2048 failed the scratch boot" with MORE free than that. Compute buffers
are allocated at graph capture during boot, so a failure is loud and fast;
the step-down rungs (4096/1536, then 4096/1024) are named in the env.
The seam stops hard-coding a rung: serve_argv now carries BATCH_SIZE and
UBATCH_SIZE the way it already carried CTX_SIZE et al., and the pair test
asserts the argv follows the shipped env plus the llama.cpp invariant
ubatch <= batch. Verified both ways: retuning the env to 8192/4096 keeps
the seam green, inverting the pair to 2048/4096 turns it red.
The prefix-caching OFF/ON experiment was confounded (both arms ran on
the marginal speed); the align-mode re-seed theory is unproven, not
disproven. Caching ships ON with zero gdn_hip faults under heavy
re-send traffic on stable RAM; knob stays commented. Corrects the
env's 'RAM RULED OUT' claim and adds the AGENTS.md lesson.
Frank: still corrupted at ~40k with --no-enable-prefix-caching deployed
and mechanically verified (zero gdn_hip faults in the container serving
that window, coherent 40,771-token boundary probe). The recurrence is
FAULT-FREE, so the align-mode page re-seed is not the (only) mechanism
and the mitigation bought whole-history re-prefill per turn for nothing.
Revert: knob FRANK_PRIMARY_NO_PREFIX_CACHING commented out -> the serve
branch renders --enable-prefix-caching (the shipped contract). The knob,
the branch and the seam stay documented for a future fixed fork build.
Open hypotheses recorded in the runbook: re-fed poisoned transcript (the
same conversation cannot distinguish fixed from unfixed), or a second
fault-free long-context path — fp8 KV at depth is the next suspect. Next
diagnostic: fresh-conversation reproduction, then a KV-dtype A/B.
Two bash defects in the 9d8b185 prefix-cache wiring, both measured live
on the 2026-10-05 deploy attempt (unit crash-looped 190x, exit 127,
endpoint down for the whole 3600 s health window):
1. A # comment block inside the backslash-continued podman run
terminates the continuation: the comment swallows the joined line
and the flag line then runs as its own command
('--no-enable-prefix-caching: command not found'). The comment moves
above the run block, with a no-comments-in-chain note.
2. The ${VAR:+A}${VAR:--B} double-expansion is ambiguous: when the
variable is set, ${VAR:--B} parses as a SUBSTRING OFFSET, not the
:- default — measured render was '--no-enable-prefix-caching1'.
The flag is now branched explicitly into
FRANK_PRIMARY_PREFIX_CACHE_FLAG (exactly one of the two flags
reaches the engine; unset keeps the historical flag).
Seam: the render test sources the shipped env through the branch and
asserts the exact flag word the engine receives (render-assert, the
same pattern the spec knobs use). bash -n clean; all 22 frank-primary
seams pass.
Measured 2026-10-04 18:46:07 PDT (live podman logs): 48 gdn_hip state
faults in one 0.6 s burst — 'state out of ssm-cache range: max|S|=0 with
non-finite values (slot=42 head=3|8)', one per TP rank, 23 GDN layers
each — then the process-lifetime report cap. The guard's own text in
this build names it: 'known Qwen3.8-Flash-Next (QFN) flaw under prefix
caching … Mitigation: --no-enable-prefix-caching'.
Mechanism: prefix caching puts QFN on mamba_cache_mode 'align', which
checkpoints each sequence's GDN recurrent state into a mamba page per
3136-token block and re-seeds it from that page when a request resumes
from a cached prefix. The NaN state landed at the 13th boundary
(13 x 3136 = 40,768 — the reported ~40k mark) and persisted in the page:
every later turn of that conversation re-seeded the same non-finite
state for two heads, silently (fault-report cap already spent).
RAM ruled out the same day: zero MCE/EDAC records since the boot under
test, 8 GiB in-band pattern stable under live load, 4800 MT/s is the
JEDEC floor on 5600-rated DIMMs, and the two-head single-slot NaN
signature is a state-path fault, not scattered bit corruption.
Fix: FRANK_PRIMARY_NO_PREFIX_CACHING=1 -> --no-enable-prefix-caching
(flag renders in both directions from the knob). Cost: every turn
re-prefills the whole history instead of at most 3135 tail tokens.
Seam updated: the prefix-cache pin now asserts the conditional render
and the shipped OFF value. Runbook carries the evidence + acceptance
criteria; all 368 tests pass.
The host region was the ONE arena-plan term still autofitting per boot: with
no cap the planner binds it to MemAvailable AT BOOT minus PLE/process/api/
margin terms (boot 2026-10-05: 237.97 GiB -> 159.39 GiB usable; the same host
reads 75.4 GiB available while running). Demand is 105.48 GiB
(52.74/rank x 2 ranks, 23,040 swap-target experts); a boot where co-tenant
RAM squeezed the bound near that silently pins next-hottest experts VRAM-only
(pinned_extra warning, smaller swap region, no flag change). 120 GiB =
measured demand + 14% headroom on 246.1 GiB total. Semantics verified in the
shipped image: cap binds as the bound, exceedance only warns.
The GPU-side arena is unchanged (deterministic arithmetic over the same
constants): verify after boot that host reads '--expert-offload-mem' and
slots stay 7,762. Seam: red first (wiring + value), then green; full suite
368+1 green.
Verified in the shipped image (v1/worker/gpu/cudagraph_mem.py):
estimate_cudagraph_memory() checks VLLM_CUDAGRAPH_MEM_ESTIMATE_MB FIRST and
returns that MiB total outright, ignoring the BASE/PER_GRAPH pair — the only
way to state the CUDA-graph plan term directly instead of deriving it. We
keep shipping the pair (0/12) so a future spec-decode shape re-prices
automatically; documented the switch condition.
Also corrected two stale shape-G claims in the runbook that predated the
1.00 revert: the deployed shape block and the lever table both still said
util 1.0 / 'le=1 ceiling', while shape G ships 0.982.
The freed third slot funded exact V; the remaining ~3.5 GiB of honest
nvidia-smi free (3,571, not the 4,046 total-minus-used I quoted - the
driver holds ~450 MiB of reserve) funds the +875 MiB of Q6_K weights.
Projected landing ~2.7 GiB free, inside the 2-3 GiB floor; measure on
boot, step the pool one 12,288-token increment if it lands under 2 GiB.
- payload dir MiMo-Q6K/ made self-contained (mmproj hardlinked from
MiMo-Q5KM/, link-count-2 verified)
- env/serve/deployer/bats repointed; download script now fetches and
verifies Q6_K (7,793,710,624 B, ef96d05a...) with Q5_K_M documented as
the rollback; payload sha verified live on the host before deploy
- seam green 14/14 against the stub engine
Record the live numbers and the UTIL 1.0 verdict (unbootable: worker init
demands free > budget; ValueError on both ranks before a single weight loads).
Driver used 30.97 of 31.86 GiB; the 0.89 GiB/card margin held through a live
4,096-token generation at ~38-39 tok/s.
Worker init refuses budget >= device total on BOTH ranks:
ValueError: Free memory on device cuda:0 (31.29/31.86 GiB ...)
NRestarts=2 crash loop, zero weights loaded. Experts stay at the v1 shape
(graph term 0.07 GiB, KV 3.1665e9 = the 400,700-token ask); only the budget
returns under the law. Arena 17.78 GiB / ~7,912 slots predicted.
Two plan terms were still unspent: the CUDA-graph charge (shape F's own boot
reported a NEGATIVE capture delta, -0.06 GiB, so the 256 MiB base pool was
fiction -> charge 0 + 12 MiB/graph = 0.07 GiB) and the budget (util 0.982 -> 1.0,
the field's le=1 maximum). Together +0.82 GiB of expert cache: plan target
18.35 GiB / ~8,016 slots. KV_BYTES 3.15e9 -> 3.1665e9 = the 400,700-token ask at
the engine's measured 7,891.8 B/token. Cost of the force, recorded: driver
high-water ~30.4 GiB of 31.86, so ~1.5 GiB of real margin per card; rollback is
UTIL=0.982.
The arena is budget - non-expert - KV reserve - graphs - overhead. With util
capped at 0.982 (never 1.00) the graph estimate is the only plan term this repo
can shrink: the fork charges an analytic 256 MiB + 52 MiB/graph = 0.55 GiB where
the engine's own post-capture line reports 0.31 GiB, so pricing it at
256 + 12/graph returns ~0.24 GiB to the expert arena. MNBT 1024 -> 2048 is
arena-free: the registry overhead is base + 0.22 GiB per 1024 ABOVE its nbt 4096
anchor, and the fitted 5.25 GiB was fitted AT 4096.
Frank's order, two messages: "extend the max context of frank-small to
200700 per individual stream without further quantization", then
"reduce it to 2 streams for a total of 401400-ish tokens, set ubatch to
1k", then "unquantize the KV if we have extra space".
CTX_SIZE 540672 -> 401408 with PARALLEL 3 -> 2 gives n_ctx_slot 200,704:
above the 200,700 the harness compacts at, so a full-length request
lands in one stream instead of a 400 exceed_context_size_error. The pool
must divide evenly across slots and stay 256-aligned for the FA path,
which is why it is 401,408 and not 401,400.
CACHE_TYPE_V q8_0 -> f16: the freed third slot is exactly what pays for
unquantized values (24.5 -> 32 KiB/token), and the budget lands back in
the 2-3 GiB band Frank set on 2026-10-02 instead of leaving ~6 GiB of
the card idle. The asymmetric-KV contract that created this endpoint is
spent with it; both --cache-type flags stay explicit.
--batch-size/--ubatch-size are now passed (2048/1024). The endpoint had
never passed either, so it ran on engine defaults nobody wrote down.
Franksdank-Runbook: serverAppConfigs-docs/LLM/frank-small-llama-mimo-q5km.md
Frank's order: replace frank-large with
kingjones777/Qwen3.8-27B-Uncensored-ROCmFP4-STRIX-MTP-GGUF's
Qwen3.8-27B-Uncensored-Q6_0_ROCMFPX_AGENT.gguf, following that card.
The file is ftype 114 — 339 GGML_TYPE_Q6_0_ROCMFPX (102) + 159
GGML_TYPE_Q8_0_ROCMFPX (103) + 353 F32 over 851 tensors,
qwen35/64 layers/full_attention_interval 4/4 KV heads/262,144 window —
read out of its own header. The pinned ROCmFPX build already knows both
types (ggml.h 102/103 plus convert/cpy/getrows/mmq/fattn device paths
for 102), so no rebuild: the fork stays at 36e9acd and the payload
directory is the only new host state.
Card compliance, all already our flags: Q4_0 draft head (not Q8_0),
--spec-draft-n-max 4 (the knee on both cards), --spec-draft-ngl 99,
--jinja, --fit off, -fa on, -dio, one slot at full window, thinking with
the card's own sampling triple (which the file also embeds as its
general.sampling defaults). Kept from the deployed shape: the BF16
projector (this repo ships only a Q8_0 one; BF16 is the file verified
4/4 on spatial ground truth here) and the byte-identical MTP head, which
the new script hardlinks instead of re-fetching 1.6 GiB.
Franksdank-Runbook: serverAppConfigs-docs/LLM/frank-large-rocmfpx-q6agent-uncensored-mtp.md
Four harness gaps that block an honest MTP re-measurement:
- measure: AR_TEMPS sweeps sampling temperatures (default "0,0.7"). Draft
ACCEPTANCE is sampling-dependent — a greedy draft matches a greedy target
exactly and is rejected more under a non-greedy target — so judging a spec
arm only at temperature 0 judges half the contract. The unsuffixed keys stay
temperature 0, so every historical number remains comparable.
- window: capture an evidence.log sidecar (expert-plan lines, compile path,
graph sizes, /metrics spec-decode and cache counters, OOM signatures). The
arena/overhead/reserve numbers exist ONLY in the boot log; an arm that loses
decode speed is otherwise unexplainable afterwards.
- orchestrator: FRANK_PRIMARY_AR_ENV_FILE stages an env other than the repo's,
which is how arms A/B without mutating the checkout (the fingerprint covers
whatever is staged, so the memo can never answer one arm with another's
numbers), plus METRIC lines for the swept temperatures and the evidence dump.
- scripts/autoresearch/frank-primary-arms.sh: runs a sequence of full windows
(baseline, MTP3+probabilistic with a step-4 capture ladder at precise memory
ON, the same at the default overhead, and the vendor SEQS-16 shape), gating
each staged env on its JSON knobs still parsing AFTER sourcing — an unquoted
{"temperature":0.7} loses its quotes at source time and reaches the engine as
{temperature:0.7}, which is a 20-minute boot wasted.
Three engine knobs that were impossible to set without editing the run line:
FRANK_PRIMARY_SPEC (--speculative-config), FRANK_PRIMARY_COMPCONFIG
(--compilation-config) and FRANK_PRIMARY_GEN_CONFIG
(--override-generation-config). Each holds the whole JSON value; empty or
unset means the flag is absent, so the shipped shape E argv is unchanged.
The seam re-pins the law that justified the omission: "speculative decode is
ABSENT BY MEASUREMENT" compared the wrong pair — c1248d4 retired MTP by
weighing fp8+MTP4 (37.4-40.6 t/s, 54% acceptance) against the bf16 no-spec
boot (36.4-39.4), bf16 then proved unbootable, and fp8 came back WITHOUT MTP.
fp8+spec was never measured against fp8+no-spec on one shape, the no-spec
incumbent runs ~30 t/s under the fleet's 35 t/s target, and every vendor
flash-next recipe ships MTP3.
Also in the seam: a dead assertion is deleted (a failing `! grep` mid-test is
exempt from bats' errexit, so the pre-precise-memory `! grep
--expert-precise-memory` never ran), and the knob test asserts by RENDERING the
flag from the sourced env instead of grepping for an assignment.
Live after the 1ac4d48 deploy: 'expert_precise_memory': True, runtime overhead
5.25 GiB, expert cache space 17.22 GiB / 7522 slots, swap region 13.7 GiB /
5986 slots, GPU KV cache 398,658 tokens / 1.99x, pool 3,150,000,000, MNBT 1024
-- the probe's projection is now production.
Compile cache proven on consecutive boots of that shape: cold 'Compiling a
graph for compile range (1, 1024) takes 179.24 s' / 'torch.compile took
188.26 s'; warm 'Directly load the compiled graph(s) ... from the cache, took
2.143 s' / 'torch.compile took 3.97 s in total' / 'init engine ... took 41.13 s
(compilation: 3.98 s)'. Host dir 1.4 GB / 4,707 files. Time to /health stays
payload-bound (490 s warm), so the claim is ~184 s of inductor/AOT build per
restart, not a proportional boot cut.
Tests: 367/367.
Measured by boot probe on the shape-D base (production stopped, candidate
booted AS the service), plan numbers read off the engine's own log:
runtime overhead 7.60 -> 5.25 GiB (precise gfx1201 TP2 registry)
expert cache 14.87 -> 17.22 GiB
slots 6,496 -> 7,522 (+1,026, +16%)
swap region 13.7 GiB / 5,986 slots
boot healthy in 420 s, no OOM
'--expert-precise-memory': True also appears in non-default args, so a silent
argparse drop is visible. The pool is NOT cut (option A) because the pool is
not an arena lever: 3.15e9/rank keeps shape D's measured 1.99x concurrency,
and a 2.2e9 arm probed in the same window (277,686 tokens, 1.38x) is recorded
as the conservative fallback.
Boot-cost bug found while diffing Frank's 3xR9700 recipe: the image ships no
/root/.cache at all, our boot writes 1.1 GB of vLLM AOT/torch_compile
artifacts (4,665 files) plus 229 MB of comgr JIT kernels into the ephemeral
layer, and podman rm -f discards them — so every deploy/restart/reboot re-pays
'torch.compile took 181.20 s in total' (173.19 s of it one graph). Mounting
FRANK_PRIMARY_CACHE_DIR at /root/.cache persists both and hides nothing.
GPU_MAX_HW_QUEUES=2 added at Frank's direction (real HSA knob); not A/B
measured. Retracted my own recipe-diff claim that cudagraph_capture_sizes
needed pinning: capture is 3 PIECEWISE + 2 FULL graphs, 3 s, 0.31 GiB.
Tests: 367/367.
deploy_plex gated on `curl http://127.0.0.1:32400/identity`, which can never
succeed: Plex accepts the TCP connection (it arrives on the container's eth0)
and then answers with nothing, so curl reports "Empty reply from server" (52).
Measured 2026-10-02 on a plex that was healthy the whole time —
`https://127.0.0.1:32400/identity` returned 200 in 44 ms while the gate burned
its full 600 s and failed the deploy.
The repo already knew this law in another place: tests/caddyfile.bats pins the
plex vhosts to a TLS upstream with the comment "PMS drops plain HTTP
off-loopback". The deployer's gate never got the same treatment, and plex's
per-service revision had been behind long enough that the gate had not run
against this posture until now.
-k because PMS serves its own self-signed cert on 32400.
Comments/docs only - no functional change. The live boot of c769ff2 reported:
GPU KV cache size: 398,658 tokens,
Maximum concurrency for 200,704 tokens per request: 1.99x
expert cache space 14.87 GiB / 6,496 slots of 2.344 MiB
The arena prediction (6,496) landed exactly; the concurrency prediction (2.04x)
did not, because the per-token KV price is SHAPE-DEPENDENT: 7,701 B/token at
MAXLEN 262144 (6.1e9 -> 792,069 tokens) but 7,892 at 200704 (3.146e9 ->
398,658). Never extrapolate a concurrency figure across geometries.
Also smoke-verified: three concurrent 67-token requests answered in parallel,
~4.1 s each - the 3 slots are real, and the pool admits 2 full-length.
Re-applies the desired shape incrementally on the proven-good fp8 3x262144
base, after bisecting why the first attempt (30904ba) crash-looped.
Root cause: `--max-num-seqs 2` cannot compile on this model. AOT
build_guards(strict_error=True) inside profile_run -> _dummy_run specializes
`query_start_loc`'s width to SEQS + 1 (3 at SEQS 2, 4 at SEQS 3) and the
ConstraintViolationError kills engine init ~7 min in. MNBT is orthogonal:
SEQS 2 failed at MNBT 1024 AND 2048; SEQS 3 passed at both. KV dtype was a
red herring. Measured by four single-variable boot probes.
Arena lever measured: the expert arena is sized from `max_model_len` via the
registry-priced KV reserve (9190 B/token x max_model_len), NOT from
--kv-cache-memory-bytes. 262144 -> 14.30 GiB / 6,248 slots; 200704 ->
14.87 GiB / 6,496 slots. The 6.1e9 -> 3.15e9 pool cut bought zero arena and
is a runtime-headroom lever.
Shape: MAXLEN 200704, SEQS 3 (unchanged - the only slot count that compiles),
KV_BYTES 3.15e9/rank = 2.04x full-length concurrency at 7,701 B/token fp8,
MNBT 1024, gather 32x64, no spec decode, no prefetch, no precise-memory.
The 2 x 200704 fp8 / MNBT 1024 shape (30904ba) crash-loops at boot: the
engine dies in determine_available_memory -> profile_run -> _dummy_run with
the SAME torch.compile dynamo constraint that killed the bf16 2 x 262144
shape on 2026-10-01:
RuntimeError: Worker failed with error 'Constraints violated
(L['query_start_loc'].size()[0])! You marked it as dynamic but your code
specialized it to be a constant (3)'
Not an OOM. Restoring the last shape measured serving (fp8 3 x 262144 /
KV 6.1e9 / MNBT 2048 / arena 6,248) so the endpoint serves again; the
desired edits are re-applied one at a time from this base.
The 2026-10-01 bf16 shape (2 x 262144, --expert-precise-memory)
crash-looped at graph capture — the arena is DEVICE-resident and that
shape had 0.13 GiB to spare. The plain bf16 2 x 262144 also did not stay
up: 0 bytes free at runtime, and a real client request (995-token prompt,
max_tokens=261149) killed the worker with .
The correction is a pool cut, not a downgrade. FP8 KV (7,701 B/token),
two 200704 windows (401,408 tokens -> 3.091e9 B floor, pinned 3.15e9),
MNBT 1024. The fork's expert planner sizes the arena AFTER the KV
override, so the ~2.95 GiB/rank reclaimed becomes arena: ~7,500 slots
vs 6,248. Two contexts at 200704 is a capacity contract, not a
throughput one (2-way concurrency summed BELOW parity at 4064 ctx), so
the slot buys the VRAM headroom long generations need.
Refs [[dankserver-qwen38-fp8-expert-offload-tuning]].
- scripts/frank-primary.env: MAXLEN 200704, SEQS 2, KV_BYTES 3.15e9,
MNBT 1024, rewritten shape comment
- scripts/serve-frank-primary.sh: header comment updated
- tests/frank_primary_unit.bats: assertions updated, new seam pins the
arena-floor pool law (the pool must move with a context change or the
reduction buys nothing)
- docs + AGENTS.md: fleet topology and deployed shape corrected
Benchmark entrypoint: bash autoresearch.sh
Goal: Make frank-primary faster: raise decode tok/s on the tcclaviger vLLM endpoint (dankserver :7770, Qwen3.8-Flash-Next-Uncensored-FP8, 2x R9700, automatic tiered-expert offload). Context size may be altered (MAXLEN/SEQS/KV_BYTES and friends); quantization may NOT be altered (the 131-shard FP8 payload and the digest-pinned image are frozen).
Runbook, both indexes, AGENTS.md (topology row, endpoint section, three new
Key Lessons, one stale SGLang label in the bench list) and the bundle log now
describe the deployed shape: pool 540,672, 180,224 tokens per conversation,
f16 K + q8_0 V, mmproj vision, 21,684 used / 2,441 free of 24,576 MiB, 106.8
tok/s decode, 3,814 t/s prefill.
The claims that mattered:
- three contexts buy 180,224 tokens per conversation, not the native 262,144,
and a longer prompt is a hard 400 - so models.yml contextWindow is 180224
and frank-primary's 183,501-token compaction request no longer fits here
(forward pointer added to the retired SGLang runbook, which asserted it)
- the vision tower bills ~1.4 GiB for a 918 MB file, established by subtracting
two measured boots, not by a formula
- /v1/chat/completions ignores the legacy top-level 'images' field: the model
invents a color; only the image_url content part reaches it
Measured after the first 3-slot deploy: 3 x 196,608 booted (n_slots=3,
n_ctx_slot=196608, multimodal loaded) but used 22,972/24,576 MiB - 1,153 MiB
free, outside the 2-3 GiB band. The two measurements now recorded in
frank-small.env price the vision tower at ~1.4 GiB (non-KV 7,390 -> 8,859
MiB), so the pool drops 49,152 tokens (16,384/slot, 1,176 MiB) to land ~2.7
GiB free. The bats budget gate is recalibrated to the measured 9 GiB non-KV
cost, which now rejects the 3 x 196,608 shape instead of blessing it.
Frank
Comrade Chief Designer