GGUF quantizations of zai-org/GLM-5.3-Flash, made with llama.cpp. 320B total parameters, 18B active. 45 layers with hybrid attention: 34 KDA linear-attention layers interleaved with 11 DSA sparse-attention layers, built on MLA, wrapped in Manifold-Constrained Hyper-Connections (mHC). 288 MoE experts with top-8 routing plus a shared expert. Context length up to 1M. Quantized from the official BF16 weights. Every tier is imatrix-calibrated and ships as 15 shards. The master row is not a file in this repo. It is listed so the numbers above have a reference point — and this time the BF16 master did fit on the machine used for the PPL measurements, so the tiers are compared against a real baseline. The layers that would hurt most under low-bit compression are protected (measured: on IQ4_XS this costs a few GiB over the bare tier): With 288 experts the expert layers dominate the file, so protecting everything else is cheap. One NextN (MTP) layer exists in the checkpoint but is excluded at conversion time via —no-mtp; these files carry the main model’s tensors only. Do not pass -ngl manually. llama.cpp fits layers to free VRAM by itself, and any explicit -ngl value — including 0 —…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: 6block
Теги: gguf, llama.cpp, glm, moe, quantized, en, zh, endpoints_compatible
Лайков: 4 | Загрузок: 2,016
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
