返回全部动态
llama.cpp b12192:mtmd 支持 MageVL-Cowork 模型
原标题:b12192
AI 摘要
llama.cpp 发布 b12192 版本,其多模态模块 mtmd 新增对 MageVL-Cowork 模型的支持。改动将 magevl 的 rope_freqs 保留在模型层级由各层共享,并在加载时校验其为 F32、元素数为 d_head 且可拆分为图所需的 4:6:6 比例。该版本同时提供 macOS、Linux、Windows、Android 等多平台预编译包。
以上摘要由 AI 生成,可能存在误差。事实请以原文为准。
正文节选
<details open> mtmd: support MageVL-Cowork model (#31819) * mtmd: support MageVL-Cowork model * mtmd: keep magevl rope_freqs at model level and check its shape The rope factors are shared by every layer, so they live in clip_model instead of a pointer copied into each clip_layer. The loader now checks that v.rope_freqs is F32 with d_head elements and that d_head splits into the 4:6:6 pairs the graph expects. --------- Co-authored-by: Pascal <admin@serveurperso.com> </details> **Website:*
发布时间:2026-11-08 05:28
抓取时间:2026-11-08 05:45
来源机构:ggml-org