# Qwen3.8-27B UD-Q8_K_XL Speculative-Decoding Matrix

This page records the speculative-decoding benchmark and the production
tuning for the unsloth Qwen3.8-27B `UD-Q8_K_XL` service on Dankclaw
(`frank-default-large`).

## Result (applied)

- Model: `Qwen3.8-27B-UD-Q8_K_XL.gguf` (29.29 GiB, bundled MTP).
- Spec: `draft-mtp`, `--spec-draft-n-max 5`, no `--spec-draft-p-min`.
- Batch 8192 / ubatch 256, parallel 2, ctx 524288 (262144 per slot).
- Realistic-prompt generation: ~13.3 t/s single, ~8 t/s per slot under
  2 concurrent streams. No premature stop.

## How MTP works on Qwen3.8

The unsloth `UD-*` GGUFs keep the MTP (Multi-Token Prediction) head inside
the model file. llama.cpp b10411+ auto-detects it and creates the draft
context from the target model. Do not pass `--spec-draft-model`.

The stock llama.cpp build must be b10411 or newer. The 2026-08-14 rebuild
on Dankclaw is b10433 (`9b05354ec`) with `GGML_HIP` + `gfx1151` + Vulkan.

## Benchmark Method

The matrix ran 54 configurations: 3 spec types (`draft-mtp,ngram-mod`,
`draft-mtp`, `ngram-mod`) x `--spec-draft-n-max` 2..10 x
`--spec-draft-p-min` {off, 0.8}. Each config loaded `llama-server`,
then ran 3 generations. The tool is `scripts/spec-matrix-bench.sh`;
analysis is `scripts/analyze-spec-matrix.py`.

## Matrix Caveat: Benchmark Prompt

The matrix used a repetitive 1100-token prompt
(`scripts/bench_live.sh` style). MTP drafts that repeated sentence almost
perfectly, which inflated the numbers for long drafts. The matrix reported
`draft-mtp` n_max=10 at ~30 t/s. A realistic story prompt gave very
different results:

| n_max | median t/s (realistic) |
| ----- | ----------------------- |
| 4 | 12.5 |
| 5 | 13.3 (peak) |
| 6 | 10.6 |
| 7 | 10.1 |
| 10 | 8.6 |
| none (baseline) | 7.2 |

Long drafts collapse to ~18% acceptance on real writing. n_max=5 balances
acceptance against draft cost.

## Premature Stop Investigation

Two failure patterns appeared during early production tests:

1. Speed collapse. `ubatch >= 512` reduced MTP acceptance to ~25% and
   generation to ~9 t/s. `ubatch 128` with `batch >= 2048` triggered a
   premature instant-EOS under 2 concurrent slots. `ubatch 256` avoids
   both: ~30 t/s single, concurrent-safe.
2. Apparent premature-EOS. The repetitive benchmark prompt with `temp 1.0`
   sometimes stops after 1-2 tokens because the model legitimately emits an
   EOS on a wall of repeated text. This is sampling variance, not a
   corruption bug. Realistic prompts never stop early.

## Production Configuration

`scripts/frank-default-large.env`:

```ini
SPEC_TYPE=draft-mtp
SPEC_DRAFT_N_MAX=5
SPEC_DRAFT_P_MIN=
BATCH_SIZE=8192
UBATCH_SIZE=256
CTX_SIZE=524288
PARALLEL=2
```

Window: 524288 total / 2 = 262144 per slot, matching the pi
`models.json` contextWindow for `frank-default-large`.

## See Also

- [[frank-default-large-benchmark]]
- [[ling-3.0-flash-trial]]
- [[deployment]]