TTQ (TokenTinkerer Quant) is an eleven-rung GGUF ladder for GrootTeam/Groot-19B-A3B-261023, from 3.56 bpw to 8.50 bpw (7.88 → 18.59 GiB), plus a Q80 vision projector for the model’s 27-block ViT. Every rung uses our TTQ recipe — the same per-tensor shape we built for the Qwen3.6-35B-A3B TTQ ladder and then replayed onto the Cyber-Tiel TTQ ladder, now applied to Groot by exact tensor name — because Groot and Cyber-Tiel share the identical 753-tensor name set. Every rung was verified against the expected per-tensor map from our Cyber-Tiel build: 10/10 exact**. Every number below is a measurement on this box, not an estimate. Upstream positions Groot as «a family of open-weight reasoning models for complex real-world tasks … compact and easy to deploy, strong on reasoning and tool-using workloads, and efficient with token budgets.» That is the model this ladder is quantizing, and the benchmark table below reads best against that claim: it is not a general-knowledge model, and the one knowledge benchmark we ran shows exactly where the compactness was paid for. > The source is better than any TTQ set has had before. These rungs are quantized directly from > a BF16 conversion of the…
Модальности:
Генерация текста Компьютерное зрение
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: TokenTinkerer
Теги: llama.cpp, gguf, quantized, imatrix, moe, qwen35moe, mtp, vision
Лайков: 3 | Загрузок: 3,439
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
