cfg-0135
backfilled
citable URL: https://halobench.com/records/cfg-0135/ — this address never moves; the anchor /records/#cfg-0135 keeps resolving
| model | Qwen3.6-35B-A3B-MTP-UD-Q4_K_M.gguf · UD-Q4_K_M · rev 0b21525e972670ed59e1812e170b27c26355381f0656ecc4e25617ece7dac58b |
| engine | ggml-org/llama.cpp 7077abbe14c510cb829c93a1328c2815b5805ebd · ROCm0/gfx1151 · host aihydra (igpu) |
| flags | HO-009 MTP n_max=3 Stage A fingerprint: same as cfg-0132 plus native model MTP/NextN path with --spec-draft-n-max 3. No ngram/external drafter, no KV quant, no proxy/llama-swap, no n_max>=4. |
| template | not recorded at test time |
| tree | upstream — stock |
aged evidence — reconstructed from the archive. Stage A activation/safety arm only. Positive draft counters were admitted, but n3 was not promoted because n2 was faster/cleaner; 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-0527 | 2026-08-20 | performance | stageA-c32768 | decode_tps=73.509404811552 |