PReader on-device models

Mirror of model files the PReader app downloads.

Hy-MT2-1.8B-STQ1_0.gguf

• Source: tencent/Hy-MT2-1.8B-1.25Bit-GGUF, file Hy-MT2-1.8B-1.25Bit.gguf, sha256 cc497fe8f033b52b3b8b00a7669e9661435432f9d4cd43f7ed24400c01507a93.
• Licence: Apache-2.0, Copyright (C) 2026 Tencent (LICENSE.txt).
• Modification: the ggml type id of the 224 STQ1_0 tensors is changed from 42 to 43 (the id llama.cpp PR #22836 assigns; 42 is Q2_0 in mainline) and general.file_type from 41 to 42. Tensor bytes are unchanged. Produced by stq1_0_gguf.py retag.
• sha256 e482a38ceaaf8420573483c96ddc8449922b5f5de6a8023b70316e65d41e6de7, 461860800 bytes.