docs: frank-secondary at 2x114688/ubatch 1024 — runbook renamed + rewritten, fleet docs updated

- runbook git mv ...-mtp-131072.md -> ...-mtp-114688.md; new Measured
  section (22,374 used / 1,751 MiB free, decode 58.2/66.9 t/s, confounded
  vs 2x131072 noted), rollback ladder, trail entry
- AGENTS.md: topology row, ACTIVE section, contextWindow 114688, runbook
  index; mmproj name corrected to the publisher's mmproj-BF16.gguf
- LLM/index.md, index.md, mimo runbook link, log.md entry + path fix

AGENTS.md

@@ -27,7 +27,7 @@ merge absorbed the former `vllmScripts` repo, history preserved).
 |--------|---------|---------|-------|------|
 | Dankserver | `frank-primary` | **radiance** (StillDeadcode v1.0.7 line; image `stilldeadcode/radiance@sha256:4c3f484564…` digest-pinned INDEX digest, 287 MiB, `libr4d` gfx1201 kernels, plain `podman run`, root systemd unit) | `Qwen/Qwen3.8-Flash-Next-FP8` official (172.8 GB, 131 shards, sha256-verified) converted **no-recipe** to `qwen4exp-fp8-native.rad` (177,109 MiB measured = 173.0 GiB) — **native modelopt block-FP8, no requantisation** — on **BOTH R9700s TP=2** (`--tp-wire wht6`), experts in a **40 GiB/rank pinned host pool** + VRAM cache (`--placement expert_tiered`, `--expert-vs-cache-ratio 0.82` = the 99 %-hit point), **PLE 47.7 GiB pinned to host RAM** (`--ngram-placement ram`; root unit + `LimitMEMLOCK=infinity`), **fp8 KV 7.32 KiB/token**, **MTP spec decode depth 3**, *…
 | Dankclaw | `frank-large` | **halogen-flash-server 0.16.2** (`ghcr.io/peonist-ai/halogen-flash-server@sha256:0c61bf84…`, digest-pinned, plain `podman run` via `scripts/serve-frank-large.sh`, root systemd unit) | Quat3rnion/halogen-swift1.5-qwen3.8-flash-next-v2-abliterated **`qwen38-flash-next-v2-swift15-abliterated.hgn`** (66,687,678,464 B = compacted r2 of the card's 89,775,868,032 B original, ~62 GiB resident; the v2 `.hgn` format loads in NO other engine), **+ REQUIRED `qwen38-flash-next-ngram.hgn`** (51,200,246,144 B, peonist @5cc17cea — read beside the checkpoint) **+ vision tower** (897,916,416 B, @e053f488 — **vision ON**, `image_url` parts only, ≥256 px probes), **MTP spec decode ALWAYS-ON (`HALOGEN_SPEC_ADAPT=0` — the adaptive default disables the head per request and HALVES decode)**, MTP head INSIDE the checkpoint (`HALOGEN_MTP_HEAD` unset; no sidecar fetched), **2 slots × 200,192 ctx** (KV pool 400,384 positions ≈ 10.7 GiB at ~28 KiB/position; the 27B slots>1-disables-drafting law does NOT carry over — batching and speculation coexist, the card's quickstart ships 2 with the head), **server-side sampler 1.0/0.95/20** (card recipe; models.yml extraBody matches), coupled cap/timeout 65536/7200, prompt cache ON by default (warm 100k turn ≈ 2 s vs ≈ 88 s cold) | 7770 |
-| Dankclaw | `frank-secondary` | llama.cpp master (0.5.0-dev, commit `5fc4f3c`), CUDA sm_86 build (root systemd unit, relinked) | ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-Dynamic-MTP-GGUF **RCO-IQ3_XXS -mtp** (27B dense qwen3_5 hybrid, 10,442,827,392 B; a per-tensor allocation mix of stock types; the MTP head in the file **drafted** — **spec ON** (draft-mtp depth 3) since 2026-10-06 evening, ngram skipped on the recurrent-rollback law) + the publisher's own **BF16 mmproj** for **vision**, on the RTX 3090, **2 × 131,072** (Frank's 128k cap each; the tightest fit in the endpoint's history — **489 MiB measured free**, the 2-3 GiB floor knowingly violated) with **q8_0 K + q8_0 V (FP8 class, 34 KiB/token)** under flash attention, batch 4096 / ubatch 512; **63.3 t/s recipe decode** (MTP acceptance 49.8 %; 48.0 no-spec). Supersedes the same-morning IQ3_S-mtp payloads: the two-slot no-spec shape and the one-slot MTP shape (both 2026-10-06; payload on disk).|7771|
+| Dankclaw | `frank-secondary` | llama.cpp master (0.5.0-dev, commit `5fc4f3c`), CUDA sm_86 build (root systemd unit, relinked) | ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-Dynamic-MTP-GGUF **RCO-IQ3_XXS -mtp** (27B dense qwen3_5 hybrid, 10,442,827,392 B; a per-tensor allocation mix of stock types; the MTP head in the file **drafted** — **spec ON** (draft-mtp depth 3) since 2026-10-06 evening, ngram skipped on the recurrent-rollback law) + the publisher's own **BF16 mmproj** for **vision**, on the RTX 3090, **2 × 114,688** (Frank's 112k cap each since the 2026-10-06 late reshape, cut from 131,072 to fund **ubatch 1024** — **1,751 MiB measured free**) with **q8_0 K + q8_0 V (FP8 class, 34 KiB/token)** under flash attention, batch 4096 / ubatch 1024; **58.2 t/s recipe decode / 66.9 (temp 0)** (MTP acceptance 42.8–54.4 %). Renamed from `frank-small` 2026-10-06 (87721ab). Supersedes the same-morning IQ3_S-mtp payloads: the two-slot no-spec shape and the one-slot MTP shape (both 2026-10-06; payload on disk).|7771|
 
 Retired endpoints (`frank-npu`, and every prior runtime) are in
 the [runbook index](#runbook-index). The full swap history lives in
@@ -308,11 +308,11 @@ image]`) — `frank-large` returned to vision with the flash-server v2 swap
 27-layer Qwen-VL tower — patching, pad-token expansion and 3-part rotary
 are the engine core; `--mm-max-patches` available),
 and `frank-secondary` through
-`mmproj-MiMo-V2.6-Distill-Qwen-9B-f16.gguf` passed as `--mmproj` — the MiMo
+the publisher's own `mmproj-BF16.gguf` passed as `--mmproj` — the Swift
 GGUF itself carries no vision tensors. `frank-secondary`'s `contextWindow` is
-**200704** (resliced from 180224 on 2026-10-03), the per-slot window, not the
+**114688** (cut from 131072 on 2026-10-06 late to fund ubatch 1024), the
-model's native 262144: the harness must compact before llama.cpp answers a
+per-slot window, not the model's native 262144: the harness must compact
-longer prompt with a 400. Note also
+before llama.cpp answers a longer prompt with a 400. Note also
 that
 the reasoning channel differs by runtime: llama.cpp AND SGLang (qwen3
 parser, live-verified) emit `reasoning_content`, vLLM emits `reasoning` —
@@ -423,8 +423,10 @@ the publisher's own `mmproj-BF16.gguf`** (931,145,888 B, sha256
 `19acd7fc…`), both `/home/dank/models/Swift15-UncIQ3XXS/`, on port 7771.
 Cutover 2026-10-05 (evening) on Frank's explicit contract: **128k
 context cap, KV at FP8, vision on, as many 128k contexts as fit**;
-reshaped twice on 2026-10-06 — morning to two 131,072 slots no-spec,
+reshaped three times on 2026-10-06 — morning to two 131,072 slots
-evening to this UNCERSED RCO-IQ3_XXS payload with MTP back on. The
+no-spec, evening to this UNCERSED RCO-IQ3_XXS payload with MTP back
+on, late the RENAME to **frank-secondary** (87721ab) at two 114,688
+slots with ubatch 1024. The
 same-day payloads it supersedes: the LIVE MiMo-V2.6 Q6_K shape and the
 Ornith-1.5-9B Q8_0 swap that was committed (77decb5..380aafa) but never
 deployed (blocked at the docs gate; the docs debt is what this section
@@ -438,31 +440,36 @@ at q8_0**. The DeltaNet state is a fixed per-slot buffer (~150 MiB/slot at
 48 linear layers × 48 value heads). Header facts were read from the Hub
 file by HTTP range + struct parse before anything was downloaded.
 
-**The shape (2026-10-06, Frank's tight-fit order, MTP back on in the
+**The shape (2026-10-06 late; Frank's tight-fit order, MTP back on in
-evening):** `--ctx-size 262144 --parallel 2` = **TWO 131,072-token
+the evening, per-slot cut + ubatch raised late):** `--ctx-size 229376
-conversations** (Frank's 128k cap each; `n_ctx_slot = CTX_SIZE /
+--parallel 2` = **TWO 114,688-token conversations** (Frank's 112k cap
-PARALLEL`), `--flash-attn on` pinned, **q8_0 K + q8_0 V** (llama.cpp's
+each; `n_ctx_slot = CTX_SIZE / PARALLEL`; cut from 131,072 to fund
+ubatch 1024), `--flash-attn on` pinned, **q8_0 K + q8_0 V** (llama.cpp's
 FP8 class — no true fp8 KV dtype exists; 8.5 effective bits; supersedes
 the same-day f16/f16 rule written for the Ornith 9B fit), all layers on
-the GPU, `--mmproj` for vision, batch 4096 / ubatch 512. **Speculative
+the GPU, `--mmproj` for vision, batch 4096 / ubatch 1024. **Speculative
 decoding is ON** (draft-mtp, depth 3): the -mtp file's head is drafted
 against the main model's own weights; ngram-mod skipped on purpose — on
 a qwen35 hybrid ngram disables recurrent-state rollback, so a rejected
 draft corrupts the DeltaNet state.
 
-**Measured at the two-slot MTP boot (2026-10-06 evening): 23,636 used /
+**Measured at the late boot (2026-10-06, 2 × 114,688 / ubatch 1024):
-489 MiB free** (`nvidia-smi` free field) — the tightest fit in the
+22,374 used / 1,751 MiB free** (`nvidia-smi` free field) — the per-slot
-endpoint's history, under the old 2-3 GiB floor BY ORDER; a 512×512
+cut freed ~1.05 GiB of KV, which paid for the ubatch-1024 buffers and
-solid-red image still answers `Red` afterwards (489 MiB free after the
+netted ~1.25 GiB back over the evening's tightest-ever fit (23,636 used
-probe), so the fit carries the vision transients. The MTP draft ctx
+/ 489 MiB free at 2 × 131,072 / ubatch 512, under the old 2-3 GiB floor
-costs **~2.4 GiB measured** (23,636 − 22,728 + 1,561) — ~6× the ~0.4 GiB
+BY ORDER; a 512×512 solid-red image still answered `Red` there, so the
-the one-slot era assumed, because the drafter's KV is built against the
+fit carries the vision transients). The MTP draft ctx costs **~2.4 GiB
-target's full 262,144 context (the draft window), not the 3-token
+measured** (23,636 − 22,728 + 1,561 at the 262,144 draft window) — ~6×
-proposal depth; the lighter RCO-IQ3_XXS quant (−1.56 GiB vs IQ3_S-mtp)
+the ~0.4 GiB the one-slot era assumed, because the drafter's KV is
-funds it. **Decode: 63.3 tok/s (recipe temp 1.0) / 72.9 (temp 0)** with
+built against the target's FULL context, not the 3-token proposal
-MTP acceptance 49.8–62.4 % (48.0 no-spec — the drafter earned its
+depth; the lighter RCO-IQ3_XXS quant (−1.56 GiB vs IQ3_S-mtp) funds it.
-2.4 GiB). `reasoning_effort` probe unchanged: `xhigh` 200, `high` → 500
+**Decode at the late boot: 58.2 tok/s (recipe temp 1.0) / 66.9 (temp
-Jinja raise (the models.yml fold stays mandatory).
+0)**, acceptance 42.8–54.4 % (the 2 × 131,072 boot read 63.3 / 72.9 at
+49.8–62.4 % — acceptance is text-dependent and the prompts differ: the
+comparison is confounded, not a shape verdict). `reasoning_effort`
+probe unchanged: `xhigh` 200, `high` → 500 Jinja raise (the models.yml
+fold stays mandatory).
 
 **Vision with the publisher's own projector**: ajgazin's repo SHIPS
 `mmproj-BF16.gguf` in the same repo (the IQ3_S era's unsloth mmproj was
@@ -485,7 +492,7 @@ payload is the UNCECERSED Swift 1.5 (23/100 refusals vs Swift 1.5's
 98/100, KL 0.0884).
 
 Runbook:
-`serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-131072.md`.
+`serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688.md`.
 Rollback: `/home/dank/models/Swift15-IQ3S/` (the IQ3_S-mtp payload live
 until the 2026-10-06 evening swap — revert da36fa3 for the two-slot
 no-spec shape) or `/home/dank/models/MiMo-Q6K/` (revert to the
@@ -1014,7 +1021,7 @@ Active:
 | Service | Runbook |
 |---------|---------|
 | frank-large (flash-server 0.16.2, abliterated Swift1.5 v2 .hgn + MTP always-on, 2×200,192, vision ON, Dankclaw) | `serverAppConfigs-docs/LLM/frank-large-flash-server-v2-swift15.md` |
-| frank-secondary (llama.cpp CUDA, Swift-1.5-27B Uncensored RCO IQ3_XXS-mtp, 2×131,072 draft-mtp q8_0 KV, Dankclaw) | `serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-131072.md` |
+| frank-secondary (llama.cpp CUDA, Swift-1.5-27B Uncensored RCO IQ3_XXS-mtp, 2×114,688 draft-mtp q8_0 KV, Dankclaw) | `serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688.md` |
 | frank-primary (radiance StillDeadcode, Qwen3.8-Flash-Next-FP8 native, 2×R9700 TP=2, Dankserver) | `serverAppConfigs-docs/LLM/frank-primary-radiance-qwen38-native-fp8.md` |
 | Dankclaw kernel posture (`amd_iommu=off`) | `serverAppConfigs-docs/LLM/dankclaw-iommu-translation-off.md` |
 | Deployment flow + incidents | `serverAppConfigs-docs/LLM/deployment.md` |
@@ -1028,7 +1035,7 @@ Retired (rollback targets and history):
 | tcclaviger vLLM fork Qwen3.8-Flash-Next-Uncensored FP8, TP=2, shape G (`frank-primary`) | Dankserver | 2026-09-29 → 10-05 (rollback target — tag `frank-primary-vllm-shapeG-20261003`; trio + env lore in the tag, image in rootless storage) | `dankserver-qwen38-fp8-expert-offload-tuning.md` |
 | FreeToken PR #587 ROCm, Qwen3.8-Flash-Next-Uncensored-FP8, ONE R9700 (`frank-primary`) | Dankserver | 2026-10-05 → 10-05 (rollback target — tag `frank-primary-freetoken-pr587-20261005`; trio + Dockerfile in the tag, image `localhost/frank-primary-freetoken:pr587-77a7201` and orcarouter payload on disk) | `frank-primary-freetoken-pr587.md` |
 | llama.cpp master MiMo-V2.6-Distill-Qwen-9B Q6_K 2×200,704 f16 KV (`frank-secondary`) | Dankclaw | 2026-10-03 → 10-05 (rollback target — payload on disk; owns the ubatch-2048 A/B; the Ornith swap of 10-05 was committed but never deployed, so this was the LIVE payload Swift replaced) | `frank-secondary-llama-mimo-q6k-2x200704.md` |
-| llama.cpp Ornith-1.5-9B Q8_0, ONE 262,144 f16 KV (`frank-secondary`) | Dankclaw | 2026-10-05 (committed 77decb5→380aafa, never deployed — superseded by Swift the same evening; payload on disk) | trail in `frank-secondary-llama-swift15-unc-iq3xxs-mtp-131072.md` |
+| llama.cpp Ornith-1.5-9B Q8_0, ONE 262,144 f16 KV (`frank-secondary`) | Dankclaw | 2026-10-05 (committed 77decb5→380aafa, never deployed — superseded by Swift the same evening; payload on disk) | trail in `frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688.md` |
 | ROCmFPX llama.cpp fork Qwen3.8-27B Q8_0_ROCMFPX_AGENT + MTP (`frank-large`) | Dankclaw | 2026-10-02 → 10-03 (rollback target, payload on disk) | `frank-large-rocmfpx-q8agent-mtp.md` |
 | gufo v0.1.1 Qwen3.8-27B UD-Q8_K_XL/UD-Q4_K_XL + DFlash2 (`frank-large`) | Dankclaw | 2026-09-28 → 10-01, again 10-01 → 10-02 (payloads removed from disk 2026-10-03; rollback = re-download) | `frank-large-gufo-qwen38-27b.md` |
 | vLLM TheRock MiMo-V2.6-Distill-Qwen-9B BF16 + cross-model MTP (`frank-large`) | Dankclaw | 2026-10-01 (same-day rollback to gufo) | `frank-large-vllm-mimo-v26-bf16.md` |

serverAppConfigs-docs/LLM/frank-secondary-llama-mimo-q6k-2x200704.md

@@ -15,7 +15,7 @@ timestamp: 2026-10-03
 # frank-secondary — llama.cpp MiMo Q6_K, 2 × 200,704, f16 KV
 
 > **RETIRED 2026-10-05 (evening).** Superseded by
-> [[frank-secondary-llama-swift15-unc-iq3xxs-mtp-131072]] — the Swift-1.5-27B
+> [[frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688]] — the Swift-1.5-27B
 > RCO-IQ3_XXS-mtp swap (uncensored, MTP drafted). Two facts for the trail:
 > committed earlier the same day (77decb5..380aafa) was **never deployed** —
 > THIS MiMo shape was the live payload when Swift cut over; (2) the KV

serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688.md

@@ -1,7 +1,7 @@
 ---
 type: reference
-title: "frank-secondary on Dankclaw — llama.cpp, Swift-1.5-27B Uncensored RCO IQ3_XXS-mtp, 2 x 131072, draft-mtp, q8_0 KV"
+title: "frank-secondary on Dankclaw — llama.cpp, Swift-1.5-27B Uncensored RCO IQ3_XXS-mtp, 2 x 114688, draft-mtp, q8_0 KV"
-description: "The active frank-secondary runbook (2026-10-05 evening; MTP back ON on the lighter RCO-IQ3_XXS since 2026-10-06) — llama.cpp master (CUDA sm_86) serving ajgazin's Swift-1.5-Qwen3.8-27B Uncensored RCO-IQ3_XXS (MTP head in the file, drafted) on the RTX 3090: TWO 131,072-token conversations at Frank's 128k cap, q8_0 K + q8_0 V (llama.cpp's FP8 class), vision through the publisher's own BF16 projector, and the full two-slot + drafter arithmetic."
+description: "The active frank-secondary runbook (live since 2026-10-05 evening; renamed from frank-small 2026-10-06, per-slot lowered to 112k and ubatch raised to 1024 the same evening) — llama.cpp master (CUDA sm_86) serving ajgazin's Swift-1.5-Qwen3.8-27B Uncensored RCO-IQ3_XXS (MTP head in the file, drafted) on the RTX 3090: TWO 114,688-token conversations at Frank's 112k cap, q8_0 K + q8_0 V (llama.cpp's FP8 class), vision through the publisher's own BF16 projector, and the full two-slot + drafter arithmetic."
 tags:
   - llm
   - frank-secondary
@@ -15,26 +15,25 @@ tags:
 timestamp: 2026-10-06
 ---
 
-# frank-secondary — llama.cpp Swift-1.5-27B Uncensored RCO IQ3_XXS-mtp, 2 × 131,072, draft-mtp, q8_0 KV
+# frank-secondary — llama.cpp Swift-1.5-27B Uncensored RCO IQ3_XXS-mtp, 2 × 114,688, draft-mtp, q8_0 KV
 
-> **ACTIVE since 2026-10-05 (evening); reshaped twice on 2026-10-06.** A
+> **ACTIVE since 2026-10-05 (evening); reshaped three times on
-> 27B replaces the 9B on the same 24 GiB card, bought by Frank's explicit
+> 2026-10-06, renamed from frank-small the same night.** A 27B replaces
-> contract for the swap: context capped at 128k per conversation, KV at
+> the 9B on the same 24 GiB card: KV at FP8 class, vision on. The
-> FP8 class, vision on, "as many of those contexts as you can". The
+> morning order (drop MTP, ubatch 512, TWO 128k contexts) bought the
-> morning order ("drop mtp, skip ngram, drop ubatch to 512, let's get 2
+> second slot — 1,397 MiB free. The evening order swapped to the
-> contexts at 128k — should be a very tight fit") bought the second slot
+> UNCELERSED Swift-1.5 at RCO-IQ3_XXS and put MTP back ON — landing
-> by spending the drafter (~0.4 GiB assumed) and ubatch headroom (~0.3
+> **489 MiB free**: the MTP draft ctx costs **~2.4 GiB, not the ~0.4
-> GiB): 1,397 MiB measured free. The evening order — swap to the
+> GiB assumed** (the drafter's KV is built against the target's full
-> UNCELERSED Swift-1.5 at RCO-IQ3_XXS and turn MTP back on — landed
+> context, not the 3-token proposal depth), paid for by the lighter
-> **489 MiB free**: the lighter quant (−1.56 GiB of weights) funds the
+> quant (−1.56 GiB). The late order dropped the per-slot window
-> second slot, but the MTP draft ctx costs **~2.4 GiB measured, not the
+> 131,072 → **114,688** (Frank's 112k cap) and raised ubatch 512 →
-> ~0.4 GiB assumed** — the drafter's KV is built against the target's
+> 1024 — landing **1,751 MiB free**, back above the morning's number
-> full context (the 262,144 draft window), not the 3-token proposal
+> with the prefill knob raised. This supersedes BOTH earlier 2026-10-05
-> depth. The 2–3 GiB floor is knowingly violated, as ordered. This
+> payloads: the live MiMo-V2.6 Q6_K shape
-> supersedes BOTH earlier 2026-10-05 payloads: the live MiMo-V2.6 Q6_K
+> ([[frank-secondary-llama-mimo-q6k-2x200704]], now retired) and the
-> shape ([[frank-secondary-llama-mimo-q6k-2x200704]], now retired) and the
+> Ornith-1.5-9B Q8_0 swap that was committed (77decb5..380aafa) but
-> Ornith-1.5-9B Q8_0 swap that was committed (77decb5..380aafa) but never
+> never deployed.
-> deployed.
 
 ## Identity
 
@@ -64,11 +63,12 @@ escape hatch is the same as every payload here: revert and redeploy.
 
 ## Shape and KV contract
 
-`--ctx-size 262144 --parallel 2` — TWO conversations at Frank's 128k cap
+`--ctx-size 229376 --parallel 2` — TWO conversations at Frank's 112k
-(the GSQ publisher's own benchmark protocol ran at 131,072 per stream).
+cap (2026-10-06 late: cut from the 128k the morning/evening shapes ran
-`CTX_SIZE` stays divisible by `PARALLEL`; the seam enforces it, and the
+at, to fund ubatch 1024). `CTX_SIZE` stays divisible by `PARALLEL`; the
-per-slot number (`262144 / 2 = 131,072`) is the product contract —
+seam enforces it, and the per-slot number (`229376 / 2 = 114,688`) is
-models.yml stays at 131,072.
+the product contract — models.yml follows at 114,688 (a longer prompt
+is a 400, never a truncation).
 
 **KV at FP8 class: `q8_0` K + `q8_0` V (Frank, 2026-10-05 evening).**
 llama.cpp has no true fp8 KV dtype; q8_0 is the 8-bit class (1.0625 B/elem
@@ -81,18 +81,25 @@ pinned: quantized V requires it, and the build's
 `GGML_CUDA_FA_ALL_QUANTS=ON` (2026-09-17) makes the path real rather than
 a silent f16 fallback.
 
-**Two slots + drafter, by arithmetic and by order (2026-10-06).** The
+**Two slots + drafter, by arithmetic and by order (2026-10-06 morning /
-morning order bought the second slot by spending the drafter and ubatch
+evening, the 2 × 131,072 era).** The morning order bought the second slot
-headroom (1,397 MiB free). The evening swap put the drafter BACK and
+by spending the drafter and ubatch headroom (1,397 MiB free). The evening
-changed the arithmetic: the draft ctx is built against the target's FULL
+swap put the drafter BACK and changed the arithmetic: the draft ctx is
-context (262,144 — the draft window, not the 3-token proposal depth), so
+built against the target's FULL context (262,144 — the draft window, not
-it costs **~2.4 GiB measured** (23,636 − 22,728 + 1,561), not the ~0.4 GiB
+the 3-token proposal depth), so it costs **~2.4 GiB measured** (23,636 −
-the one-slot era assumed. The lighter quant pays for it: **23,636 used /
+22,728 + 1,561), not the ~0.4 GiB the one-slot era assumed. The lighter
-489 MiB free** (nvidia-smi `free` field) — the tightest fit in
+quant pays for it: **23,636 used / 489 MiB free** (nvidia-smi `free`
-frank-secondary's history, the 2–3 GiB floor knowingly violated, and the
+field) — the tightest fit in frank-secondary's history, the 2–3 GiB floor
-512×512 vision transient (34 MiB) still fits and recovers. Compute buffers
+knowingly violated, and the 512×512 vision transient (34 MiB) still fits
-allocate at graph capture during BOOT, so a miss is loud and fast — the
+and recovers.
-failure mode is a crash loop, not silent corruption.
+
+**The late reshape (2 × 114,688, ubatch 1024) is the current fit.** The
+per-slot cut frees 2 × 16,384 tokens × 34 KiB ≈ 1.05 GiB of KV; ubatch
+512 → 1024 spends part of it on compute buffers (a prefill knob — decode
+never moves with it). Measured boot: **22,374 used / 1,751 MiB free**.
+The drafter's own KV shrank with its target (draft window 229,376).
+Compute buffers allocate at graph capture during BOOT, so a miss is loud
+and fast — the failure mode is a crash loop, not silent corruption.
 
 ## Speculative decoding — ON since 2026-10-06 evening (draft-mtp, depth 3)
 
@@ -161,7 +168,7 @@ leave room for thinking.
 - **Vision**: 512×512 solid-red PNG → `Red` again on this shape.
 - Deploy: commit 812d9eb, e2e gate 33/33 green on the live fleet.
 
-## Measured: the 2 × 131,072 draft-mtp boot (2026-10-06 evening, live)
+## Measured: the 2 × 131,072 draft-mtp boot (2026-10-06 evening, historical)
 
 - **Boot ledger**: `creating MTP draft context against the target model
   '…/Swift-1.5-Qwen3.8-27B-Uncensored-Dynamic-MTP-RCO-IQ3_XXS.gguf'`,
@@ -186,6 +193,29 @@ leave room for thinking.
   slot prefix cache is live.
 - Deploy: commit da36fa3, e2e gate 33/33 green on the live fleet.
 
+## Measured: the 2 × 114,688 / ubatch-1024 boot (2026-10-06 late, live)
+
+- **argv** (from `/proc/<MainPID>/cmdline`): `--ctx-size 229376
+  --parallel 2` (per-slot 114,688), `--batch-size 4096 --ubatch-size
+  1024`, `--spec-type draft-mtp`. Deploys 87721ab (the rename — old
+  `frank-small.service` stopped + disabled first, by hand: `disable` on
+  a relinked unit removes only its symlink) then f2575a1 (the shape);
+  health gate + MainPID check green, e2e 32 passed / 1 flaky-retry.
+- **VRAM: 22,374 used / 1,751 MiB free** (nvidia-smi `free` field) —
+  the per-slot cut (≈1.05 GiB of KV) paid for the ubatch-1024 buffers
+  and netted back ~1.25 GiB against the evening's 489 MiB; back above
+  the morning's 1,397 MiB.
+- **Decode + MTP acceptance** (live `timings`): recipe (temp 1.0) →
+  **58.2 t/s** (430/1005, 42.8 %); temp 0 → **66.9 t/s** (475/873,
+  54.4 %). The 2 × 131,072 boot measured 63.3 / 72.9 — −8–10 %, but
+  acceptance is text-dependent and these prompts differ from that
+  measurement's, so the comparison is CONFOUNDED, not a shape verdict;
+  a clean A/B needs stop + mask (bench policy).
+- **Smoke**: `READY` completion + `reasoning_content` populated; the
+  harness reached :7771 under the new alias (models.yml provider/id
+  `frank-secondary`, config.yml modelRoles rewritten; per-slot
+  `contextWindow: 114688`).
+
 ## Client contract
 
 - Reasoning in `reasoning_content` (`--reasoning-format deepseek`);
@@ -209,14 +239,17 @@ leave room for thinking.
 
 ## Rollback
 
-The 2 × 131,072 no-spec IQ3_S-mtp shape is ONE commit away — revert
+**The immediate predecessor shape (2 × 131,072, ubatch 512, same
-da36fa3 and redeploy: the env trio then ships `SPEC_TYPE=none`,
+payload) is ONE revert away** — revert f2575a1 and redeploy
-`CTX_SIZE=262144`, `PARALLEL=2`, `UBATCH_SIZE=512` against
+(`CTX_SIZE=262144`, `UBATCH_SIZE=512`). The 2 × 131,072 no-spec
-`/home/dank/models/Swift15-IQ3S/` (the IQ3_S-mtp payload, still on disk,
+IQ3_S-mtp shape is one revert further — revert da36fa3 next: the env
-measured 48.0 tok/s no-spec). The 1 × 131,072 MTP shape is two reverts
+trio then ships `SPEC_TYPE=none`, `CTX_SIZE=262144`, `PARALLEL=2`,
-back (812d9eb's parent): the same IQ3_S-mtp file with
+`UBATCH_SIZE=512` against `/home/dank/models/Swift15-IQ3S/` (the
-`SPEC_TYPE=draft-mtp`, `UBATCH_SIZE=1024`, `CTX_SIZE=131072`, `PARALLEL=1`
+IQ3_S-mtp payload, still on disk, measured 48.0 tok/s no-spec). The
-(measured 58.9–66.9 tok/s with the head drafting).
+1 × 131,072 MTP shape is two reverts further back (812d9eb's parent):
+the same IQ3_S-mtp file with `SPEC_TYPE=draft-mtp`, `UBATCH_SIZE=1024`,
+`CTX_SIZE=131072`, `PARALLEL=1` (measured 58.9–66.9 tok/s with the head
+drafting).
 
 Three payloads on disk, all fetchable by the same script's git-history
 recipes:
@@ -260,3 +293,13 @@ recipes:
   target's full 262,144 context, not the 3-token proposal depth) plus the
   second 131,072 slot: 489 MiB free, vision `Red` at a 128-token budget,
   decode 63.3 t/s recipe / 72.9 temp 0 (48.0 no-spec).
+- 2026-10-06 (late, Frank): the endpoint renamed frank-small →
+  **frank-secondary** (87721ab — scripts, unit, seam, runbooks,
+  identifiers; historical references to the OLD FreeToken/CUDA
+  frank-small, and to the Dankserver service that became frank-medium,
+  stay frank-small) and the per-slot window cut 131,072 → 114,688 to
+  fund ubatch 512 → 1024 (f2575a1). Live unit transition done by hand
+  before the deploy: `systemctl stop frank-small` + `systemctl disable
+  frank-small`, then `deploy.sh dankclaw --service frank-secondary`.
+  1,751 MiB free; decode 58.2 t/s recipe / 66.9 temp 0; models.yml +
+  config.yml retargeted.

serverAppConfigs-docs/LLM/index.md

@@ -103,18 +103,20 @@ frank-secondary, port 7771]
   (`slow`/`advisor` → frank-primary) carried forward. Superseded by the
   MODEL swap: the abliterated v2 .hgn loads only in flash-server, so the
   engine moved with the model
-- [[frank-secondary-llama-swift15-unc-iq3xxs-mtp-131072]] — **ACTIVE since
+- [[frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688]] — **ACTIVE since
   2026-10-05 (evening); UNCECERSED RCO-IQ3_XXS with MTP back on since
-  2026-10-06 evening** (root systemd unit, port 7771): llama.cpp master
+  2026-10-06 evening; RENAMED from frank-small and reshaped to 2 ×
+  114,688 / ubatch 1024 late 2026-10-06** (root systemd unit, port
+  7771): llama.cpp master
   (CUDA sm_86) serving ajgazin/Swift-1.5-Qwen3.8-27B-Uncensored-Dynamic-MTP-
   GGUF **RCO-IQ3_XXS-mtp** (27B; the MTP head in the file is DRAFTED —
   draft-mtp depth 3, ngram skipped on the recurrent-rollback law) on the
-  RTX 3090, TWO 131,072-token conversations at Frank's 128k cap (pool
+  RTX 3090, TWO 114,688-token conversations at Frank's 112k cap (pool
-  262144 / parallel 2 — the tightest fit in the endpoint's history:
+  229376 / parallel 2 — cut from 131,072 to fund ubatch 1024; **1,751
-  **489 MiB measured free**, the 2-3 GiB floor knowingly violated) with
+  MiB measured free**) with
-  q8_0 K + q8_0 V (FP8 class, 34 KiB/token vs 64 at f16), ubatch 512, and
+  q8_0 K + q8_0 V (FP8 class, 34 KiB/token vs 64 at f16) and
-  vision through the publisher's own mmproj-BF16.gguf. Decode 63.3 t/s
+  vision through the publisher's own mmproj-BF16.gguf. Decode 58.2 t/s
-  recipe / 72.9 temp 0 (MTP acceptance 49.8–62.4 %; 48.0 no-spec). The
+  recipe / 66.9 temp 0 (MTP acceptance 42.8–54.4 %).
   RCO-IQ3_XXS tier is a per-tensor allocation mix of stock types; the
   payload is the UNCECERSED Swift 1.5 (23/100 refusals vs Swift 1.5's
   98/100). Replaces BOTH same-day payloads: the live MiMo Q6_K shape

serverAppConfigs-docs/index.md

@@ -372,14 +372,15 @@ The LLM inference fleet on the two hosts. Full index:
 - [[ciru-activefpx-promptforge]] — the RETIRED v3 CIRU PromptForge runtime on Dankclaw (superseded 2026-08-24)
 - [[kingjones-strix-mtp]] — the kingjones Qwen3.8-27B ROCmFP4 STRIX MTP model on Dankclaw (active)
 - [[dealign-ornith-crack]] — the retired Ornith-1.5-35B-A3B CRACK model on Dankserver (retired 2026-09-02)
-- [[frank-secondary-llama-swift15-unc-iq3xxs-mtp-131072]] — the ACTIVE
+- [[frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688]] — the ACTIVE
   frank-secondary runbook since 2026-10-05 (evening; **UNCECERSED RCO-IQ3_XXS
-  with MTP back on since 2026-10-06 evening**): llama.cpp master (CUDA
+  with MTP back on since 2026-10-06 evening; renamed from frank-small and
+  reshaped to TWO 114,688-token conversations at Frank's 112k cap with
+  ubatch 1024 late 2026-10-06**): llama.cpp master (CUDA
   sm_86) serving ajgazin's Swift-1.5-Qwen3.8-27B Uncensored RCO-IQ3_XXS
-  (the MTP head in the file is drafted) on the RTX 3090 — TWO
+  (the MTP head in the file is drafted) on the RTX 3090 — 1,751 MiB
-  131,072-token conversations at Frank's 128k cap, the tightest fit in
+  measured free, q8_0 K + q8_0 V, vision
-  the endpoint's history (489 MiB measured free), q8_0 K + q8_0 V, vision
+  through the publisher's own projector, ~58 t/s recipe decode with MTP
-  through the publisher's own projector, ~63 t/s recipe decode with MTP
 - [[frank-secondary-llama-mimo-q6k-2x200704]] — RETIRED 2026-10-05 (evening,
   superseded by swift15): llama.cpp master serving
   bartowski/MiMo-V2.6-Distill-Qwen-9B-GGUF **Q6_K** — TWO conversations at

serverAppConfigs-docs/log.md

@@ -2,6 +2,26 @@
 
 
 ## 2026-10-06
+- **frank-secondary (Dankclaw): RENAMED from frank-small, then reshaped
+  to 2×114,688 / ubatch 1024 (llama.cpp, RTX 3090).** The rename (commit
+  87721ab, 87 files: serve trio, unit, bats seam, 9 runbooks, identifiers
+  `FRANK_SECONDARY_*`/`llm_secondary`/`deploy_llm_secondary`; historical
+  refs to the OLD FreeToken/CUDA frank-small and to the Dankserver
+  service that became frank-medium stay frank-small, as does
+  `freetoken-frank-small.md`). Live unit transition by hand before the
+  deploy: `systemctl stop frank-small` + `systemctl disable frank-small`
+  (a relinked unit — disable removes only its stale symlink) →
+  `deploy.sh dankclaw --service frank-secondary`. The reshape (commit
+  f2575a1, Frank's order "lower max context to 112k so we can increase
+  ubatch to at least 1k"): `CTX_SIZE 262144→229376` (per-slot
+  131,072→114,688), `UBATCH_SIZE 512→1024` → **22,374 used / 1,751 MiB
+  free** (the per-slot cut frees ~1.05 GiB of KV; ubatch 1024 spends
+  part, netting ~1.25 GiB back over the 489 MiB fit), decode 58.2 t/s
+  recipe / 66.9 temp 0 (acceptance 42.8–54.4 % — the drop vs the prior
+  boot's 63.3/72.9 is confounded by prompt text, not a shape verdict).
+  Harness retargeted: models.yml (id/name/compactionModels,
+  `contextWindow: 114688`) + config.yml modelRoles ×14. Runbook renamed:
+  `serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688.md`.
 - **frank-secondary (Dankclaw): two Swift-1.5-27B reshapes (llama.cpp, RTX 3090).**
   Morning: Frank's order "drop mtp, skip ngram, ubatch 512, two 128k
   contexts" → 2×131,072 no-spec on the IQ3_S-mtp (commit 812d9eb):
@@ -19,7 +39,7 @@
   uncensored model thinks more — 64 tokens is all `reasoning_content`).
   ngram-mod still skipped (recurrent-state rollback law on the qwen35
   hybrid). Runbook renamed + rewritten:
-  `serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-131072.md`.
+  `serverAppConfigs-docs/LLM/frank-secondary-llama-swift15-unc-iq3xxs-mtp-114688.md`.
 - **frank-large (Dankclaw): halogen 27B-dense → halogen-flash-server
   0.16.2 v2 Swift1.5 ABLITERATED (Strix Halo iGPU, port 7770).** Cutover
   commit 31b1ec3 (03:07:40): the v2 `.hgn` payload loads ONLY in the