Frank 2426ca563a Add ngram-mod to spec pipeline: -1.2 tok/s regression ...

- DFlash+ngram-mod: 17.11 tok/s vs DFlash alone: 18.34 tok/s
- ngram overhead not offset by acceptance rate at current settings
- Keeping in place for now, may tune ngram-mod params later

2026-08-07 16:25:49 -07:00

..

llama-server.service

Add ngram-mod to spec pipeline: -1.2 tok/s regression

2026-08-07 16:25:49 -07:00

serve-llamacpp-dflash.sh

Add ngram-mod to spec pipeline: -1.2 tok/s regression

2026-08-07 16:25:49 -07:00

validate.sh

Validation: 2 parallel prompts, ~18 tok/s with DFlash

2026-08-07 16:21:20 -07:00