# Lane A — Affordability report (tic 748) → your inbox, tic 751

**The report itself:** `lane-A-affordability-report-tic748.json` (this folder; a byte-copy of `audit-logs/governance/hoist-preparation-tic748/lane-A-affordability.json`, 86 KB, measured 2026-08-28T19:20:40Z, authored by {'standing': 'task_scoped_worker', 'agent_type': 'general-purpose', 'dispatched_by': 'ent_homeskillet', 'model_id': 'claude-opus-5[1m]', 'inscription_authority': False, 'outputs_owned_by': 'ent_homeskillet'}).

**Its one-line summary (verbatim):** Thirteen T7 entries, ~181 GiB of models (plus 35G tmux-dumps); the library is 4x what the boot packet names. The GGUF lane (27B at 17.0 GiB resident, 7.31 tok/s measured; Qwythos-9B at 8.9 GiB, 120 receipts) is FULLY INTACT and costs nothing to light. The MLX lane is DARK — ~/.cache/huggingface is gone, so the epoch16 proposer's 8.1 GiB base is absent and five receipts point at nothing; one download relights it. The 35B AgentWorld converts to ~20.0 GiB resident (fits the measured 24.96 GiB Metal budget alone) but its ~18.5 GiB output has nowhere internal to land — 15 GiB free — making the biggest hoist a DISK and MEMBRANE question, not a RAM one, and the coupling itself the Architect's bell.

## Hardware as it measured it (tic 748)
```
{
 "model": "Mac14,5",
 "cpu": "Apple M2 Max",
 "ncpu": 12,
 "unified_memory_bytes": 34359738368,
 "unified_memory_gib": 32.0,
 "unified_memory_source": "sysctl -n hw.memsize -> 34359738368",
 "os": "macOS 27.0 (sysctl kern.osproductversion)",
 "metal_gpu_working_set_mib_measured": 25559,
 "metal_gpu_working_set_gib_measured": 24.96,
 "metal_gpu_working_set_source": "llama.cpp device_info line in /Volumes/T7 Shield/models/model-mlx-maybe/tasks/bjz958hid.output: 'MTL0 : Apple M2 Max (25559 MiB, 25558 MiB free)' — the runtime's own report of the device budget at load, on this machine, with the 27B GGUF. This is the REAL GPU-resident ceiling, not a rule of thumb.",
 "iogpu_wired_limit_mb_NOW": 0,
 "iogpu_wired_limit_source": "sysctl iogpu.wired_limit_mb -> 0 (= default/unset)",
 "iogpu_wired_limit_note": "The tic-594 raise to 28672 (SOVEREIGN-FLEET-RUNBOOK.md §0, sudo, non-persistent across reboot) is NO LONGER IN EFFECT. Every co-residency claim resting on a 28 GiB cap ('both engines fit') is currently FALSE-BY-STATE. Single-model budget is the default ~24.96 GiB.",
 "disk": {
  "T7_Shield": {
   "size": "3.6Ti",
   "used": "3.2Ti",
   "avail_gib": 407,
   "capacity": "90%",
   "device": "/dev/disk6s2",
   "source": "df -h '/Volumes/T7 Shield'"
  },
  "internal_data_volume": {
   "size": "926Gi",
   "used": "872Gi",
   "avail_gib": 15,
   "capacity": "99%",
   "device": "/dev/dis
```

## Affordability verdicts (tic 748)
```
{
 "formula_declared": {
  "resident_bytes": "weights_bytes + kv_cache_bytes + runtime_overhead",
  "weights_bytes": "params_total * bits_effective / 8",
  "bits_effective": {
   "MLX 4-bit (group 64)": "4.5  (4 bits/weight + one fp16 scale + one fp16 bias per 64 weights => 4 + 32/64)",
   "GGUF Q4_K_M": "~4.85",
   "GGUF Q6_K": "~6.56",
   "GGUF Q8_0": "~8.5",
   "bf16": "16"
  },
  "kv_cache_bytes_per_token": "2 (K and V) * n_kv_heads * head_dim * n_FULL_ATTENTION_layers * bytes_per_element",
  "kv_note": "For hybrid linear/full-attention models only the FULL-attention layers carry a per-token KV cache; the linear-attention (GatedDeltaNet) layers carry a CONSTANT-size recurrent state instead.",
  "headroom_rule": "The binding ceiling on this machine is the Metal GPU working set MEASURED at 25,559 MiB (24.96 GiB) with iogpu.wired_limit_mb at its default 0, NOT the 32 GiB of unified memory and NOT the dispatch's '~8 GiB headroom' heuristic. The heuristic and the measurement happen to land close (32 - 8 = 24 vs 24.96 measured); I use the measured one."
 },
 "ranking": [
  {
   "rank": 1,
   "candidate": "Qwen3.6-27B-MTP Q4_K_M GGUF (llama.cpp Metal, :8088)",
   "verdict": "FITS-NOW",
   "params_total": "27B",
   "params_active": "27B (dense)",
   "quantization_on_disk": "Q4_K_M + MTP draft head",
   "disk_on_hand_gib": 15.93,
   "disk_needed_gib": 0,
   "resident_gib_estimate": 17.0,
   "resident_breakdown": "16.0 GiB weights (27B * 4.85/8) + 0.51 GiB MTP draft context (llama.cpp's OWN estimate in the log: 'estimated memory usage of MTP context is 521.00 MiB') + ~0.3 GiB KV @ 4096 ctx + ~0.2 GiB buffers",
   "fits_32gib_m2max": true,
   "fits_under_measured_metal_budget_24_96gib": true,
   "co_resident_with_a_second_engine": "NOT at the current default cap. 17.0 + 11.1 (org-engine server) = 28.1 GiB > 24.96 GiB. The tic-594 runbook's 'both fit' finding required sudo sysctl iogpu.wired_limit_mb=28672, which is NOT currently set.",
   "measured_receipts_on_this_machine": [
    "tic 544 first-fire: 162 tokens (60 prompt + 102 gen), 7.31 tok/s gen, MTP 43.6% draft-acceptance, load ~71 s, 127.0.0.1 only, clean shutdown verified (audit-logs/governance/receipts/2026-07-
```

## Inventory headline
```
[
 {
  "entry": "organization-engine-lora",
  "du_measured": "20G",
  "index_size": "20.0 GiB",
  "index_agrees": true,
  "what_it_is": "The org-engine LoRA epoch archive. 252 files. Composition MEASURED by du: 14 epoch .zip adapters at ~456M each (epoch05,06x2,08,09,10,11,12,13,14,15,16 + the epoch05 eval 128K) = ~6.0G; the 2.7G compact-continuation export (org_engine_epoch_lora_Qwen3-14B-unsloth-bnb-4bit_compact_continue_20260616T220931Z.zip, sha 15fe6578...); upload_chunks_20260617 2.7G; self_foreclosure_upload_staging 457M; training/ 9.0G (additive_v1 8.9G incl. remote_training_v1/runs; additive_v2 23M — DATA ONLY, no weights); receipts/ 384K (one file: epoch15_f2_governed_tic514_receipt.json); README.md 3515 B.",
  "canonical_head": "org_engine_epoch16_s6_f2_rebalance_from_epoch14_20260626T191807Z_sdpa.zip (478,023,981 B), sha256 b7db58dac68326901679daa14624d0e607c91438dd268476f9549
```

## CORRECTIONS from tic 751 — read these beside the report
1. **The internal-disk figure in this report is `df`'s purgeable-EXCLUSIVE free.** APFS container free reads 14.4 GB, but the OS-reported available (purgeable-inclusive) is the operative headroom (~65 GB by your reading); the OS reclaims purgeable space on demand. A floor written on the `df` number (my recommendation at /review 750, ratified) blocked the lap-2 measurement for two tics on the wrong quantity. RETRACTED at tic 751 on your word (ruling_751b on the covenant object).
2. **`~/.ollama` holds 15 GB of model blobs on the INTERNAL disk that nothing uses** — the engine is llama.cpp serving from the T7 (`ollama ps` empty at every reading); `ollama-library` already exists on the T7. Not placed this tic; which tic pulled them is not recorded here. Your files: removing or moving them frees 15 GB.
3. **The models live on the T7 and are served from it** — the 749 lap ran the 27B GGUF directly from `/Volumes/T7 Shield/models/` (mmap; nothing copied inward). The only real LOCAL limit is RAM: `iogpu.wired_limit_mb=0` → one engine resident at a time; the lap's legs were always sequential, so this never required two resident models.
4. **The router was not used for the laps** — neither 749 nor 751 routed through canonical-mount `invoke` (the one live verb); the local leg went through the AK providers row, the frontier probe through raw `claude -p`. Any cloud-hosted leg belongs behind the router.
5. **Tic 751 status:** one-model lap in flight (the 27B off the T7, one leg, the frontier leg declared not_run), per your direction.