cfg-0137
backfilledfork
citable URL: https://halobench.com/records/cfg-0137/ — this address never moves; the anchor /records/#cfg-0137 keeps resolving
| model | Qwen3.8-27B-Q8_0.gguf · Q8_0 · rev 57484e3196aaff8dbbd71c666158b00206baa2bc134e4f16ae90fc8cbdfbeea2 |
| engine | ggml-org/llama.cpp 0b0f35d0ed745bd6e40eb248e320cacbb1a546ab · Vulkan0/Strix-Halo · host aihydra (igpu) |
| flags | HO-004 stock MTP n_max=3 Stage A fingerprint: same target/runtime/backend/KV/batch/load-mode as cfg-0136 plus native draft-mtp path with --spec-type draft-mtp --spec-draft-n-max 3. No DFlash2 drafter, no ngram/proxy/llama-swap/ROCm, no stock MTP n_max>=4, no alternate target quant, no KV quant substitution. |
| template | not recorded at test time |
| tree | fork — carrying con-0002 |
aged evidence — reconstructed from the archive. Stage A activation/safety arm only. Positive draft acceptance counters were reviewer-admitted, but the arm failed the Stage A sentinel gate on code/toolish prompts; no guard or Stage B performance cell is admitted and no n_max>=4 safety claim is made.
Capability basis
measured on this config
Runs on this config (1)
| run | date | kind | suite | metrics |
|---|---|---|---|---|
| run-0542 | 2026-08-20 | guard | stageA-c32768-varied-prompts | runner_invalid=true |