HookTime AI Briefing model (Qwen2.5-0.5B, q8, .litertlm)
LoRA fine-tune of Qwen2.5-0.5B-Instruct for the HookTime: Fishing Forecast Android app.
Generates a short on-device daily fishing briefing from structured forecast data
(solunar windows, tides, wind, waves, pressure). Runs via Google LiteRT-LM
(litertlm-android 0.13.1), seq128, q8 quantization. ~492 MB, generates a ~130-word
briefing in ~25-30 s on a low-end phone (Snapdragon 480-class, CPU x4).
Not a general-purpose chat model - it is tuned for one prompt format only.
